Skip to main content

Full text of "Managing the plant-animal interface in tropical legume-grass pastures"

See other formats


MANAGING THE PLANT -ANIMAL INTERFACE IN 
TROPICAL LEGUME-GRASS PASTURES 



BY 

JORGE LUIS MORALES GONZALEZ 



A DISSERTATION PRESENTED TO THE GRADUATE 
SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL 
FULFILLMENT OF THE REQUIREMENTS FOR THE 
DEGREE OF DOCTOR OF PHILOSOPHY 



UNIVERSITY OF FLORIDA 



'UNIVERSITY 



1990 



OF FLORIDA LIBRARIES 



To my parents Jose and Zoraida 
To my wife Graciela 

To my children Jorge, Diana and Jose Alonso 
I love you all 



ACKNOWLEDGMENTS 



The author thanks Dr. John E. Moore for the privilege and the 
learning experience of working with him during the Ph.D. program. His 
understanding of animal nutrition concepts, methods and research is an 
attractive academic option to any student interested in animal science. 
Thanks are extended to Dr. William F. Brown, co-chairman of the doctoral 
program. His guidance in the conduct of the experiment, statistical 
analyses and interpretation of results is fully appreciated. Thanks go 
to Dr. Lynn E. Sollenberger for being a member of the committee and for 
his advice. Thanks go to Dr. H. H. Van Horn, chairman of the committee 
for the master's degree and now committee member in the doctoral 
program. The author feels indebted to him because he was an important 
factor in the continuation of the the Ph.D. program. Thanks also go to 
Dr. William G. Blue for being a committee member and for his instruction 
in soil science. 

Specials thanks are given to the Ministry of Agriculture and 
Livestock of Costa Rica for the extension of permission to continue the 
doctoral program immediately after the master's degree. Thanks are 
extended to all fellow students and friends who made the difficult 
student life much easier. 

Special thanks also go to the University of Florida, represented by 



iii 



the Animal Science Department, for economic support through a research 
assistantship during the Ph.D. program and for the opportunity to 
receive a quality education. Thanks go to the staff of the Ona 
Agricultural Research and Education Center for collaboration on field 
work and laboratory analyses . 

This study was supported in part by USDA Tropical Agriculture Grant 
No. 86-CRSR-2-2823 "Managing the plant-animal interface in tropical 
legume -grass pastures." 



iv 



TABLE OF CONTENTS 

PAGE 

ACKNOWLEDGMENTS iii 

LIST OF TABLES vi 

ABSTRACT x 

INTRODUCTION 1 

LITERATURE REVIEW 3 

Characteristics of Florida Livestock Production 3 

Grazing Systems 9 

Plant-Animal Interface 11 

MATERIALS AND METHODS 20 

Pastures 20 

Sampling and Laboratory Procedures 22 

Statistical Analysis 30 

RESULTS AND DISCUSSION 32 

Pasture Canopy Structure and Forage Nutritive Value 32 

Diet Botanical Composition and Nutritive Value 54 

Animal Ingestive Behavior 69 

Pasture Canopy Characteristics and Animal Ingestive 

Behavior Relationships 77 

SUMMARY AND CONCLUSIONS 86 

APPENDIX 90 

LITERATURE CITED 108 

BIOGRAPHICAL SKETCH 114 



v 



LIST OF TABLES 



TABLE PAGE 

1 Sampling dates for pastures and ingestive behavior 

measurements 23 

2 Least squares means for the effect of pasture type 

(bahiagrass (B) and bahia-aeschynomene (BA)) on whole 
canopy herbage mass, and in vitro organic matter digestion 
(n-24) 33 

3 Least squares means for the effect of grazing management 
(before (BRG) and after (ARG) rotational grazing, and 
continuous grazing) on whole canopy herbage mass, height 

and in vitro organic matter digestion (n=16) 34 

4 Least squares means for the interaction of grazing 
management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) and pasture type 
(bahiagrass (B) and bahia-aeschynomene (BA)) on whole 

canopy botanical composition, and crude protein (n=8) 36 

5 Least squares means for the effect of cycle on whole 

canopy legume percentage in bahia-aeschynomene pastures 
(n=6) and in vitro organic matter digestion in all pastures 
(n-12) 39 

6 Least squares means for the interaction of grazing 

management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) and canopy layer on 
dead material percentage across all pastures (n=12) and on 
legume leaf and stem percentage in bahia-aeschynomene 
pastures (n=6) 41 

7 Least squares means for the interaction of pasture type, 

grazing management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) and canopy layer on 
herbage mass and grass leaf percentage (n=6) 45 

8 Least squares means for the interaction of pasture type, 

grazing management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) and canopy layer on 
forage nutritive value (n=6) 47 

vi 



f 



9 

10 

11 

12 

13 

14 

15 

16 

17 

18 

19 



51 

53 

55 

57 

61 

66 

68 

70 

72 

75 

78 



Least squares means for the interaction of cycle and canopy 
layer on legume leaf and stem percentage in bahia- 
aeschynomene pastures (n-6), and on dead material percentage 
across all pastures (n-12) 

Least squares means for the interaction of cycle, pasture 
type, and canopy layer on weed percentage (n-12) 

Least squares means for the effect of pasture type 
(bahiagrass (B) and bahia-aeschynomene (BA)) on botanical 
composition and nutritive value of the diet ingested by 
animals (n-24) 

Least squares means for the effect of grazing management 
(before (BRG) and after (ARG) rotational grazing, and 
continuous grazing (CG)) on diet botanical composition 
and nutritive value (n-16) 

Least squares means for the interaction of pasture type 
(bahiagrass (B) and bahia-aeschynomene (BA)), and grazing 
management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) on diet grass leaf 
and stem percentage; and grazing management effect on diet 
legume leaf and stem in bahia-aeschynomene pastures (n-8) . 

Least squares means for the effect of cycle on diet legume 
leaf and stem percentage in bahia-aeschynomene pastures 
(n-12) 

Least squares means for the interaction of cycle and 
pasture type on diet crude protein concentration (n=6) .... 

Least squares means for the effect of pasture type 
(bahiagrass (B) and bahia-aeschynomene (BA)) on animal 
ingestive behavior (n-24) 

Least squares means for the effect of grazing management 
(before (BRG) and after (ARG) rotational grazing, and 
continuous grazing (CG)) on animal ingestive behavior 
(n-16) 

Least squares means for the interaction of pasture type and 
grazing management (before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) on bites per minute 
(n-8) 

Simple correlations between canopy height or grass 
percentage and ingestive behavior for bahiagrass pastures 
(n-12; r-correlation coefficient; P=probability value) .... 



vii 



20 

21 

22 

23 

24 

25 

26 

27 

28 

29 

30 

31 



Simple correlations between whole canopy characteristics 
and animal ingestive behavior for bahia-aeschynomene 
pastures (n-12; r-correlation coefficient; P-probability 
value) 80 

Simple correlations between canopy layer characteristics 
and animal ingestive behavior for bahia-aeschynomene 
pastures (n-12; r-correlation coefficient; P-probability 



value) 83 

Analysis of variance for whole canopy herbage mass, height 
and botanical composition (MS-mean square; P-probability 
value) 91 

Analysis of variance for whole canopy legume percentage in 
bahia-aeschynomene pastures (MS=mean square; P-probability 
value) 92 

Analysis of variance for whole canopy nutritive value 
(MS-mean square; P-probability value) 93 

Analysis of variance for canopy layer herbage mass, and 

botanical composition (MS-mean square; P-probability 

value) 94 

Analysis of variance for canopy layer botanical 

composition (MS-mean square; P-probability value) 95 

Analysis of variance for canopy layer legume fractions 
(MS-mean square; P-probability value) 96 

Analysis of variance for canopy layer nutritive value 
(MS-mean square; P-probability value) 97 

Analysis of variance for grazing horizon herbage mass and 

botanical composition (MS-mean square; P-probability 

value) 98 

Analysis of variance for grazing horizon dead material and 

forage nutritive value (MS-mean square; P-probability 

value) 99 

Analysis of variance for grazing horizon legume fractions 
(MS-mean square; P-probability value) 100 

Analysis of variance for bottom layer herbage mass and 

botanical composition (MS-mean square; P-probability 

value) 101 



viii 



33 Analysis of variance for bottom layer dead material and 
forage nutritive value (MS=means square; P-probability 

value) 102 

34 Analysis of variance for bottom layer legume fractions 

(MS-mean square; P=probability value) 103 

35 Analysis of variance for diet botanical composition 

(MS-mean square; P=probability value) 104 

36 Analysis of variance for diet legume fractions of animals 

grazing bahia-aeschynomene pastures (MS=mean square; 
P-probability value) 105 

37 Analysis of variance for diet nutritive value (MS-mean 

square; P-probability value) 106 

38 Analysis of variance for animal ingestive behavior 

(MS-mean square; P=probability value) 107 



Abstract of Dissertation Presented to the Graduate School 
of the University of Florida in Partial Fulfillment of the 
Requirements for the Degree of Doctor of Philosophy 



MANAGING THE PLANT -ANIMAL INTERFACE IN 
TROPICAL LEGUME-GRASS PASTURES 

By 

JORGE LUIS MORALES GONZALEZ 
August 1990 

Chairman: John E. Moore 

Co-Chairman: William F. Brown 

Major Department: Animal Science 

'Pensacola' bahiagrass ( Pasoalum notatum Flugge) is the most widely 
used perennial grass for summer grazing in Florida; however, animal 
performance from this grass is low. Overseeding bahiagrass (B) pastures 
with the summer annual legume Aeschvnomene americana L. may improve the 
nutritive value of B pastures but pasture -management practices may 
influence the quantity of legume in the pasture and in the diet of 
grazing cattle. Grazing management [rotational (RG) vs. continuous (CG) 
grazing] of B and bahiagrass -aeschynomene (BA) pastures was studied by 
investigating the plant-animal interface during grazing of these 
pastures by cattle. Animal defoliation effects during rotational 
grazing were studied by sampling both before (BRG) and after (ARG) 
rotational grazing. 

Overseeding B pastures with aeschynomene increased forage crude 
protein (CP), particularly in the grazing horizon. Herbage mass (HM) 



x 



was greater for BRG than for CG. Forage in vitro organic matter 
digestion (IVOMD) was greater for RG than for CG. Greater grass-leaf 
percentage in both pasture types and higher legume percentage in BA 
pastures were found for RG than for CG. In BA pastures, highest forage 
CP concentration was observed for BRG when legume percentage was 
highest. 

Diet CP concentration of animals on BA pastures was 46% higher than 
that of animals on B pastures. For RG, diet CP was reduced by animal 
defoliation so that the level for ARG was similar to that for CG. 

Animals consumed a diet higher in IVOMD during RG than CG. 

Animals on BA pastures obtained a smaller bite weight (BW) than did 
those on B pastures. Intake per minute (IPM) , however, was not 
different because presence of legume increased bites per minute (BPM) , 
which compensated for the smaller BW. Animals obtained greatest BW when 
sampling BRG. For ARG, however, BW was reduced to a level similar to 
that for CG. The IPM was greatest for BRG. A depression in BPM for 
ARG, however, depressed IPM to a level even lower than that for CG. 

Aeschynomene increases forage nutritive value of B pastures. 

Grazing BA pastures rotationally gives animals the opportunity for a 
greater intake of forage of higher nutritive value compared to 
continuous grazing. Greater intake of greater nutritive value forage, 
however, is obtained only during the first part of the rotational 
grazing period. Therefore, maximum benefit will be obtained if high- 
producing animals grazed these pastures first and they are removed to a 
new paddock before severe defoliation of the legume occurred. 



xi 



INTRODUCTION 



In Florida, bahiagrass ( Paspalum notatuxn Flugge) comprises most of 
the land in improved perennial pastures (0.90 million hectares; 

Chambliss and Jones, 1981). This warm-season perennial grass is widely 
used because of its excellent persistence under heavy grazing 
conditions. However, it is of relatively low quality and only moderate 
to low animal performance can be achieved from it. Summer grazing is 
important to Florida livestock producers in terms of calf weaning weight 
and cow body condition. It is imperative, therefore, to increase the 
quality and utilization of summer grazing pastures. Several other 
improved grass species such as stargrass ( Cvnodon nlemfuensis ') and 
limpograss ( Hemarthria altissima ) have been released over time. 

Increased amounts of forage and beef per area can be obtained from these 
new improved grasses if increasing amounts of fertilizer, particularly 
nitrogen fertilizer, are applied to them. 

An alternative to high cost fertilizers for improved tropical 
pastures has long been the association of grass pastures with legumes. 
Improved forage nutritive value and animal performance have been 
achieved when legumes have been introduced (Peacock et al., 1976; Roger 
et al. , 1983a; Kalmbacher and Martin, 1983). Aeschynomene ( Ae s chvnomene 
americana L.), a summer -growing annual legume, has been successfully 
utilized in association with different types of grasses in Florida 
(Brown et al., 1987; Sollenberger et al., 1987b; Rusland et al., 1988). 



1 



2 



Pasture nutritive value and animal performance have been improved when 
it was grown with bahiagrass (Hodges et al., 1974). Given the different 
growth habits and persistence under grazing of grasses and legumes, 
grass -legume associations may be difficult to maintain. 

Forage plants and cattle are affected by environmental variables 
such as soil, temperature and rainfall, and by grazing management. 

Plant and animal responses to these factors, and their interactions, 
make grazing a very dynamic system. Plants and plant parts of higher 
nutritive value in the pasture canopy of the pasture may not be readily 
available to the animal as a consequence of these responses and 
interactions . Knowledge in this area might indicate the appropriate 
management of tropical grass -legume pastures to improve forage 
utilization and animal performance. Plant-animal interface studies 
offer the opportunity to gain knowledge about the components of a 
grazing system and their interactions. 

Based on the above considerations, a study on managing the plant- 
animal interface in Pensacola bahiagrass and bahia-aeschynomene pastures 
was conducted with the following general objectives: a) to determine 

the effects of rotational and continuous grazing management systems on 
pasture canopy structure (herbage mass, forage botanical composition and 
forage nutritive value), diet botanical composition and nutritive value, 
and animal ingestive behavior, and b) to relate pasture canopy 
characteristics to animal ingestive behavior. 



LITERATURE REVIEW 



Characteristics of Florida Livestock Production 
Type of Production System 

Cow-calf production is the most important commercial operation in 
the Florida beef industry (Baker, 1980). Weaned calves are usually 
marketed from July to September. Summer is a very important period for 
the beef industry in Florida. During this period the rancher has 
calves, his major marketing product, to put weight on before weaning 
time. He also has lactating cows that must gain in body condition 
during summer. The summer forage growing season offers the opportunity 
to provide lower cost nutrients to these animals. 

Pasture Grazing Resources 

There are 4.9 million hectares of grassland in Florida, 1.3 million 
of which are improved perennial grass pastures (Spreen et al., 1985). 

Of these improved perennial grass pastures, over 0.90 million hectares 
are of bahiagrass ( Paspalum notatum Flugge). There are three cultivars 
of bahiagrass grown in Florida: Pensacola, Argentine and Paraguay; of 
which Pensacola is the most popular. Therefore, bahiagrass is the base 
grass used in improved pastures for summer grazing in Florida 
(Ocumpaugh, 1978). 



3 



4 



Bahiagrass 

Bahiagrass is a warm- season perennial grass forming a dense sod and 
stolon-root system. The location of the phytomer- producing meristems is 
at the base of the stolon tip. Because the tip is not removed when 
clipping or grazing, bahiagrass has a high degree of tolerance to 
defoliation (Sampaio et al. f 1976). Bahiagrass responds to N 
fertilization with increased forage production and increased forage N 
concentration (Blue, 1983). Twenty three kilograms or less of applied 
N, however, might not alter N concentration of the forage (Beaty et al., 
1960) . It has been suggested that because bahiagrass tolerates a high 
degree of defoliation its management should be directed more towards its 
quality rather than its regrowth (Beaty et al., 1968). 

The desirable agronomic characteristics of bahiagrass are probably 
responsible for its extensive use in Florida. Animal performance on 
bahiagrass pastures, however, is somewhat limited. When compared to 
other grasses like 'Callie' bermudagrass ( Cvnodon dactvlon ) . Tifton 
hybrid bermudagrass 72-81 and Tifton hybrid bermudagrass 72-84, daily 
gain was lower for Pensacola bahiagrass (0.53, 0.46, 0.52, and 0.38 kg, 
respectively; Bertrand and Dunavin, 1988). Prates et al. (1974) studied 
performance of steers on fertilized continuously- grazed Pensacola 
bahiagrass. Daily gain was 1.0 kg in May but declined to -0.52 kg in 
September. Forage crude protein (CP) concentration remained above 11% 
throughout the experiment, but in vitro organic matter digestion (IVOMD) 
tended to decline as the season progressed. It was suggested that the 
decreased gain in summer was related to a decline in forage quality and 
also to problems of rejection and waste due to higher stocking rates in 



5 



summer and resulting contamination by feces and urine. In a grazing 
study from June to September on Argentine bahiagrass, Moore et al. 

(1969) reported high forage nutritive value with respect to digestion 
coefficients for dry matter (55.6 to 59.4%) and cellulose (68.7 to 
72.7%). Forage digestible protein was inadequate for growing cattle, 
however. Digestible protein percentages measured in three different 
periods during that experiment were 3.8, 2.6 and 1.5%, respectively. 
According to a citation in the same paper (Milford and Haydock, 1965) 
digestible protein in a subtropical forage should be 3.6% to insure a 
positive nitrogen balance. Brown and Mislevy (1988) in a clipping study 
found that increasing maturity reduced CP and IVOMD of Pensacola 
bahiagrass, Pangola digitgrass ( Digitaria decumbens Stent.) and Ona 
stargrass ( Cvnodon nlemfuensis var. nlemfuensis) at very rapid rates. 
They also measured the acid detergent insoluble crude protein (ADICP) 
which is a measure of the protein that is bound to the fiber components 
of the cell wall and unavailable to the animal. As a proportion of 
total CP concentration, ADICP increased from 9.3% at 2 weeks regrowth to 
51.3% at 8 weeks regrowth during the summer harvest. Heifers (Pitzer et 
al., 1988) grazing Pensacola bahiagrass, and calves (Kunkle et al., 

1988; Kunkle and Baldwin, 1988) grazing bahiagrass (cultivar not given) 
pastures had higher daily gains when supplemented with protein. In 
spite of forage CP levels above 8.3%, animals showed low protein status 
as indicated by blood urea N. Low protein status of these animals might 
reflect low levels of available protein in the forage as observed by 
Moore et al. (1969) and Brown and Mislevy (1988). 



6 



The benefit of including legume forages in perennial pastures has 
been demonstrated by several studies in Florida. Inclusion of winter 
legumes in forage programs resulted in higher weaning percentage, 
slightly heavier calves and a lower total cost per cow (Peacock et al., 
1976; Roger et al., 1983a; Roger et al., 1983b). Blue (1983) estimated 
that the value of forage and protein in white clover (Trifolium repens ) - 
Pensacola bahiagrass pastures were equal to those produced by grass with 
a nitrogen application rate between 224 and 448 kg/(ha year) . The 
benefit of including summer legumes in perennial pastures has also been 
demonstrated in Florida (Hodges et al., 1974, 1976; Pitman, 1986; 

Rusland et al., 1988). Inclusion of summer legumes has increased gain 
per animal, gain per hectare and cow reproductive performance. 

Ocumpaugh (1978) suggested that overseeding bahiagrass pastures with the 
summer annual legume , aeschynomene ( Ae s chvnomene americana L) , should 
improve summertime gains. 

Aeschynomene 

Ae s chvnomene americana has been used in perennial grass pastures in 
Florida for many years (Hodges et al., 1982), and has been the focus of 
attention in many studies in Florida. It is the most widely adapted 
warm- season legume available for grazing in south and central Florida 
(Hodges et al., 1982; Pitman and Rretschmer, 1984). 

Aeschynomene is an annual legume capable of re-establishing itself 
from seed. Natural reseeding is not very reliable unless management 
(Chaparro, 1989) and environmental conditions are favorable. Ralmbacher 
and Martin (1983) suggested that light penetration to the base of the 
canopy and soil water relations appear to be important in promoting good 



7 



legume establishment. Recommendations for establishment of aeschynomene 
in bahiagrass sod are given by Kalmbacher et al. (1988). After 
seedlings are 2 or 3 weeks old they recommended fertilizing with 33 kg 
P 2 C> 5 /ha and 66 kg K^O/ha. Recommended seeding rate for bahiagrass -sod 
seeding of aeschynomene is 6.7 to 9.0 kg/ha of de-hulled seed. Poor 
stands of aeschynomene resulted at the Ona Agricultural Research and 
Education Center when less than the average of 30 mm rain fell during 
February to May. They recommended choosing the seeding date carefully 
as the amount of rain before and 2 weeks after seeding aeschynomene is 
related to stand establishment. Liming before seeding is important if 
soil pH is below 5.5. Aeschynomene shows tolerance to long-term periods 
of flooding, but not to soil moisture deficits (Albrecht et al., 1981). 
When seeded in limpograss pastures, initiation of grazing when 
aeschynomene was 0.2 to 0.4 m tall as compared to 0.6 and 0.8 m tall 
resulted in more uniform distribution of total and legume dry matter, 
higher efficiency of grazing, more vigorous legume regrowth, and a trend 
toward greater total herbage consumption (Sollenberger et al., 1987b). 
Hodges et al. (1982) suggested that aeschynomene should be rotationally- 
grazed because grazing before the plant attains sufficient size results 
in low productivity. 

Aeschynomene had leaves and young stems that contained 24.1% CP 
(Hodges et al., 1982); however, the coarse stems contained 6.1% CP. 
Pasture nutritive value was improved when aeschynomene was grown with 
bahiagrass (Kalmbacher and Martin, 1983). Selectivity for aeschynomene 
in the pasture and palatability of aeschynomene leaf and fine stems at 
the top of the plant were demonstrated with fistulated animals 



8 



(Sollenberger et al., 1987b). Kretschmer et al. (1986) found in a study 
with several selected aeschynomene introductions, however, that in 
September as compared to June, IVOMD of Pangola digitgrass was in many 
instances equal to or greater than that of the associated legume. 
Sollenberger et al. (1987b) showed that the major contribution of 
aeschynomene to limpograss- aeschynomene mixtures was its own CP 
concentration. 

Aeschynomene showed adaptation to Florida flatwoods in mixture with 
bahiagrass pastures under grazing (Pitman and Kretschmer, 1984). Hodges 
et al. (1974) evaluated eight forage and supplement systems in south- 
central Florida over a 5 -year period. The aeschynomene treatment, which 
consisted of 50% Pangola digitgrass and 50% of a mixture of bahiagrass, 
and aeschynomene, showed average weaned calf percent of 82% compared to 
67% for the perennial grass treatment which consisted of pure stands and 
mixtures of digitgrass, bahiagrass and common bermudagrass . Calf 
production per cow and calf production per hectare were also higher for 
the aeschynomene treatment than for the perennial grass system. 
Aeschynomene treatments produced intermittent legume grazing of 
excellent quality from August to November in some but not all years. 
Failures were due to below-normal June and July rainfall. Gaps in the 
grazing supply of aeschynomene systems required more hay feeding than on 
the perennial-grass program. They suggested that no more than 25% of a 
pasture system should be planted to aeschynomene. Rusland et al. (1988) 
found increased daily gains of steers and gain per hectare when 
aeschynomene was grown in limpograss pastures during midsummer through 
early fall. 



9 



Aeschynomene in mixture with tropical grasses improves forage 
nutritive value and animal performance during the summer season. A good 
stand of aeschynomene is necessary, however, to achieve the benefits of 
this legume in mixture with tropical grasses. A good stand of 
aeschynomene is obtained if recommendations for establishment and 
management discussed above are followed. It is very important that 
summer rains are frequent at the time of seeding because this has a 
great impact on the germination and development of seedlings and, so, on 
the future of the legume stand. 

Grazing Systems 

Grazing management is the tool used to exercise control of forage 
utilization. There are basically two systems of grazing management, 
continuous and rotational. In the continuous grazing system, animals 
have unrestricted access to a pasture. In the rotational grazing 
system, pastures are grazed intermittently. A rest period is given to 
the pasture to allow it to recuperate from defoliation, allowing 
regrowth and accumulation of herbage for a new grazing period. There 
has long been a controversy about which grazing system is superior in 
terms of animal production (Wheeler, 1962). The controversy centered on 
the role of stocking rate in determining animal production per area. 
Comparison between these two systems has been done mostly at different 
stocking rates. Rotationally grazed pastures usually have been stocked 
higher and so this gave an advantage to rotational grazing. When the 
two systems were compared at similar stocking rates, not much difference 
was apparent between systems (Wheeler, 1962). Both management systems 



10 



have advantages that might fill practical needs in specific situations. 
In some cases a combination of the two systems in different periods 
during the grazing season might be more beneficial in terms of available 
forage and liveweight gain than the use of one particular grazing system 
during the entire season (Blaser, 1982). The choice of a particular 
grazing system depends on factors such as animal type, forage species, 
period or season of the year, available resources and animal production 
obj ectives . 

Rotational -grazing systems have been proposed with several 
modifications (Blaser et al., 1986). One of these is the leader- 
follower rotational -grazing system. In this modification, animals with 
higher nutritional demands such as high producing milking cows or 
growing animals are allowed to graze the paddocks first. These cattle 
will have the opportunity to obtain the most nutritious components 
available in the pasture. Once the leader animals graze the more 
nutritious components in the canopy they are moved to a new paddock. 
Animals with lower nutritional needs then finish grazing the remaining 
available forage. Blaser (1982) showed higher milk production and daily 
gain of leader grazers as compared to follower grazers. On average, 
there might not be any advantage in milk production or daily gain by 
using this modified system compared to a straight rotational system 
(Archibald et al. , 1975; Blaser, 1982) and it might even complicate the 
management of the herd, as suggested by Archibald et al. (1975). 
Rotational grazing systems are options to improve daily gain or milk 
production of the most productive animals. Research should provide the 
most appropriate grazing system for particular situations. Studies on 



11 



plant-animal interface might provide some insight into adequate grazing 
management practices for particular cases. 



Plant-Animal Interface 



Definition 

Plant-animal interface is described as the interactions between 
forage plants and animals during the act of consumption of forages 
(Moore and Sollenberger , 1986). In a grazed pasture there are 
continuing changes in sward and animals and continuing mutual feedback 
between sward and animals (Riewe, 1980). It is necessary to study the 
individual components in a grazing system and, also, the mechanisms by 
which they interact to provide the appropriate management to optimize 
the animal output from a pasture. The two components of the plant- 
animal interface are defined under the terms of 'pasture canopy 
structure ' and 'animal ingestive behavior. ' 

Ingestive behavior 

Intake of digestible energy is a basic determinant of animal 
performance (Moore, 1981). Voluntary intake is controlled by distention 
and metabolic mechanisms. Under grazing, ingestive behavior may 
override either of the other two mechanisms (Moore and Sollenberger, 
1986). Components of animal ingestive behavior include bite weight, 
rate of biting and grazing time. The product of these three components 
should give an estimation of herbage intake of grazing animals over a 
given time interval (Hodgson, 1982) . Rate of biting and grazing time 
are used by animals as compensatory mechanisms in face of decreased bite 



12 



weight. The action of these two mechanisms in preventing a decline in 
intake is, however, limited (Hodgson, 1985). Jaw movements and bites 
per 100 jaw movements are also described as ingestive behavior 
mechanisms. Bites per 100 jaw movements indicate amount of forage 
manipulation occurring during prehension (Moore et al., 1985). Bite 
weight was shown to be the major factor influencing the intake of 
herbage by grazing cattle (Chacon and Stobbs, 1976) and animal 
production (Chacon et al., 1978). 

Ingestive behavior may be affected by pasture canopy 
characteristics (Stobbs, 1973a; Chacon et al., 1978; Moore et al., 1985; 
Brown et al. , 1987). Inaccessibility of the most nutritious components 
in the pasture canopy might reduce herbage intake (Stobbs, 1975). This 
indicates that some non-nutritional characteristics of the pasture might 
influence herbage intake (Hodgson, 1985) and animal performance (Chacon 
et al. , 1978) . 

Canopy structure 

Pasture canopy structure includes measurements such as herbage 
mass, canopy height, botanical composition, and bulk density (weight per 
unit volume) of total herbage or botanical components. The pasture can 
be characterized as a whole canopy or further by stratifying the canopy 
in layers perpendicular to the vertical plane (Stobbs, 1973b). Canopies 
vary in terms of spatial distribution of botanical components 
(Hendricksen and Minson, 1985; Sollenberger et al., 1987a). Canopy 
structure is changed by animal defoliation (Chacon and Stobbs, 1976; 
Brown et al., 1987), nitrogen fertilization (Stobbs, 1975), forage 
species (Chacon et al., 1978) and grazing management (Stobbs, 1973b). 



13 



Canopy structure influences animal ingestive behavior (Stobbs, 1973a; 
Stobbs, 1975; Chacon and Stobbs, 1976; Ludlow et al. , 1982; Moore et 
al. , 1985; Forbes and Hodgson, 1985; Brown et al., 1987), the diet 
selected by the animal (Chacon and Stobbs, 1976; Sollenberger et al., 
1987a) and the nutritive value of the herbage within the sward (Stobbs, 
1973b) . Bite weight is also the ingestive behavior mechanism most 
directly influenced by canopy characteristics, particularly by herbage 
mass and canopy height (Hodgson, 1985) . 

Plant animal- interface studies on monocultures 

Stobbs (1973a) showed the effect of sward canopy structure on bite 
weight in a series of experiments. Bite weight was calculated from the 
weight of the collected extrusa obtained with esophageally-fistulated 
animals and the number of jaw movements recorded with an automatic 
device. Pastures of fertilized (50 kg N/(ha month) during 2 years) and 
unfertilized Setaria anceps cv. Kazungula were used. Dry matter yield, 
plant height, plant bulk density (kg/(ha cm)), nitrogen percentage and 
the concentration of some minerals in the forage were higher in the 
fertilized pastures. Mean bite weight by animals grazing the fertilized 
pastures was larger than by animals grazing the unfertilized pastures 
(0.39 vs 0.13 g OM/bite). 

In a second study, Stobbs (1973a) measured bite weight by animals 
grazing 5 -week old regrowth of the legume siratro ( Macroptilium 
atropurpureum) . Kazungula setaria and Pangola digitgrass. Dry matter 
yields were similar among the three pasture types; however, plant height 
was greatest in setaria and lowest in digitgrass (26.3 vs 19.4 cm). 

Plant bulk density was greatest in digitgrass and lowest in setaria (199 



14 



vs 149 kg/(ha cm)). In vitro dry matter digestion was greatest in 
siratro and lowest in setaria (67.5 vs 59.8%). Mean bite weight was 
smaller for animals grazing siratro pastures than for animals grazing 
setaria or digitgrass pastures (0.24, 0.38 and 0.34 g OM, respectively). 

In a third experiment, pure pastures of Pangola digitgrass and 
rhodesgrass ( Chloris gavana cultivar Pioneer) were treated with the 
growth regulators gibberellic acid (GA) , and 2-chloroethyl- 
trimethylammonium chloride (CCC; Stobbs, 1973a). The denser canopy of 
both pastures after treatment with GA resulted in a greater bite weight 
than that obtained from these same forage species after treatment with 
CCC. Treatment with CCC increased herbage yield and sward height, and 
produced erect, stemmy plants with long internodes. Stobbs (1973a) 
concluded that sward bulk density, a low stem content and a high 
leaf/height ratio have a major influence upon the bite weight of cattle 
grazing these pastures. It was estimated that a mean bite weight of 
less than 0.30 g OM/bite (bite equal to jaw movement in this case) can 
seriously limit herbage daily intake of 400-kg liveweight cattle. 

Pasture maturity affected bite weight (Stobbs, 1973b). Swards of 
Qblotts gay ana and Setaria anceps at 2 , 4 , 6 , and 8 weeks of regrowth 
were compared. Animals had maximum bite weight when pastures were 
grazed at 4 -week regrowth. At this regrowth, pastures had the highest 
proportion of accessible leaf. At 2-week regrowth, pastures contained 
82% leaf, but the yield and density of the herbage was low and bite 
weight was also low. At 6- and 8 -week regrowth, pastures had large 
herbage yields and bulk densities, but animals had small bite weight. 
Negative correlations were found between bite weight and yield of 



15 



herbage and sward bulk density. Bite weight was positively correlated 
with leaf percentage in the pasture, and even more highly correlated 
with leaf percentage in the top layers of the canopy. 

In Setaria anceps pastures (Stobbs, 1975), N fertilization at 0, 50 
and 100 kg/ha increased bite weight in a linear fashion: 0.29, 0.33 and 
0.37 g OM/bite, respectively. The increase in bite weight was related 
to higher leaf yields and bulk densities. High levels of N 
fertilization promoted faster rate of growth of 6 -week regrowth pastures 
with no further increase in the bite weight of the animals. Nitrogen 
fertilization increased leaf yield and bulk densities of these pastures, 
particularly in the uppermost layers allowing animals to harvest heavy 
bites. It was suggested that pastures should be grazed before they 
become mature, not only to obtain a higher nutritive value diet but also 
to achieve a satisfactory herbage intake by means of the behavioral 
mechanism. From these studies it is apparent that sward bulk density 
positively affects bite weight. However, when the increase in sward 
bulk density is due to plant maturity the relationship becomes negative. 

Chacon and Stobbs (1976) studied the effect of progressive 
defoliation of Setaria anceps pastures on canopy structure and ingestive 
behavior. At the beginning of grazing periods, animals selected mostly 
leaf from the top of the canopy. As the quantity of leaf decreased due 
to defoliation, animals took smaller bites, increased grazing time and 
increased rate of biting. At later stages of defoliation, grazing time 
and biting rate declined. Bite weight at later stages of defoliation 
was extremely low, between 66 to 79 mg OM/bite, which resulted in a low 



16 



herbage intake. Leaf yield, leaf percentage and green material bulk 
density were major sward factors influencing intake by grazing animals. 

Swards of N- fertilized Setaria anceps and Dieitaria decumbens 
continuously stocked at 4.3 animals per hectare supported higher animal 
growth rate, as compared to swards stocked at 6.2 and 8 animals per 
hectare (Chacon et al., 1978). Steers were able to obtain a heavier 
bite and select more nutritious plant parts from the top of the sward at 
the lower stocking rate. 

Ludlow et al. (1982) studied the effect of increasing sward density 
of tropical pastures on bite weight. They also observed the effect of 
increasing sward density on forage yield as increased density might 
affect light distribution and canopy photosynthesis. Density of Setaria 
sphacelata and Dieitaria decumbens pastures were modified with the 
growth regulators GA and CCC. Results indicated that Setaria sphacelata 
and Dieitaria decumbens leaf area densities could be increased to values 
exceeding those found in temperate pastures without great reductions in 
yield. Consistent with the results of Stobbs (1973a), increased bite 
weight was obtained as sward density increased. The conclusion from 
these results was that there is considerable scope for increasing bite 
weight and possibly animal production in situations where bite weight 
(<0.30 g OM/bite) limits daily intake of forage. The researchers 
suggested also that sward density could be increased by pasture 
management because density varies with season, stocking rate and 
frequency of defoliation. 



17 



Plant-animal interface studies on grass -legume associations 

Moore et al. (1985) studied the effect of different canopy 
characteristics (canopy height and legume-grass proportions) of 
limpograss-aeschynomene pastures on animal ingestive behavior. Steers 
grazing these pastures responded to increased legume percentage in the 
upper layer by making more manipulative jaw movements and decreasing 
rate of biting, but intake per bite did not change. Intake per bite 
increased as the percentage of green herbage in the upper layer 
increased. Animals selected a diet greater in legume percentage than 
that found in the upper layer of the canopy as a result of the 
manipulative jaw movements prior to biting. Increased canopy height was 
associated with a larger bite weight. 

Brown et al. (1987) studied the plant-animal interface in stargrass 
and stargrass -aeschynomene pastures rotationally- grazed with 4-week rest 
periods. They observed a higher biting rate at the beginning of the 
grazing period with a quadratic decline at the end of the grazing 
period. They suggested that this effect was due to a reduction in 
herbage allowance as the grazing period progressed. Biting rate tended 
to be higher in the grass -legume compared to the grass -alone pastures. 

In general, differences in biting rate and diet composition were higher 
between early- and midpoint- grazing than between middle and end of the 
grazing period. Manipulations, in terms of bites per 100 jaw movements, 
increased as the grazing period progressed and tended to be higher in 
grass-alone pastures. Bite weight decreased in both types of pasture 
from beginning to end of the grazing period. Animals on grass -alone 
pastures, however, obtained higher bite weight than did those on grass- 



18 



legume pastures. The higher biting rate of animals in grass -legume 
pastures was not enough to compensate for the smaller bite weight. 
Therefore, animals on grass -legume pastures had a lower intake per 
minute. Another observation was that at the beginning of grazing 
periods animals selected for grass leaf in the grass alone -pastures . In 
grass -legume pastures, however, selection shifted towards legume leaf. 

From the plant-animal interface studies reviewed, it is clear that 
canopy structure is affected by several factors. Some of these factors 
are forage species, plant maturity, N fertilization, animal defoliation 
and grazing management. Grazing animals respond to changing canopy 
characteristics by modifying their ingestive behavior. Some canopy 
characteristics influence ingestive behavior in a more favorable way 
towards a greater herbage intake. Increased herbage intake means 
improvement in animal performance. There is an indication from these 
studies that canopies with higher bulk densities of grass and/or legume 
leaf, particularly in the upper layers, are more favorable towards 
animal ingestive behavior responses that maximize intake. Grazing 
management of forage species and grass -legume associations to produce 
those canopy characteristics that favor ingestive behavior for maximum 
intake needs to be investigated. 

The profits of the beef industry in Florida originate from heavier 
weaning weights and cows in good body condition. These two aspects of 
economic importance to Florida producers can be improved by increasing 
intake of forage with a better nutritive value during summer grazing. 
Bahiagrass is the base of improved pasturage for summer grazing in the 
state. Aeschynomene is a legume adapted to Florida grasslands that can 



19 



improve the nutritive value of bahiagrass pastures. The proper 
management of bahia-aeschynomene pastures to obtain a canopy structure 
that maximizes the voluntary intake of herbage needs to be investigated. 

It is apparent from this review that studying the plant-animal 
interface might have some practical implications. It might give some 
knowledge about the most adequate pasture and animal management 
practices for specific situations. Knowing the nutritive value and 
forage availability of a pasture is not enough to estimate animal 
performance. Study of the plant-animal interface goes a step further 
and characterizes the mechanisms and relationships between the plant and 
animal components of the system that might be important to animal 
performance . 



MATERIALS AND METHODS 



General Description of Location 



Pastures of Pensacola bahiagrass ( Paspalum notatum Flugge) and 
Pensacola bahiagrass -aeschynomene ( Ae s ch vnomene americana L.) were 
grazed from August to November, 1986 at the Ona Agricultural Research 
and Education Center near Ona, in Hardee County. The experimental site 
was located at 82° 55' W longitude and 27° 26' N latitude in south 
central Florida. Pastures were on a sandy siliceous hyperthermic Ultic 
Haplaquod (Pomona fine sand) soil. Soil pH was 5.5. Total 
precipitation for 1986 was 1393 mm, which was 36 mm above the 1942-86 
average of 1357 mm. Of this precipitation 1081 mm fell between June and 
November. The average minimum and maximum temperatures between June and 
November were 16.6 and 32.6° C, respectively (Stephenson and McCloud, 
1987) . 



Pastures 



Pasture Establishment 

Pastures were located on an established stand of bahiagrass in a 
total area of about 0.50 ha. The entire area was fertilized June 1, 
1986 with 12 kg P/ha and 46.5 kg K/ha. An area of 0.26 ha, to be used 
as pure bahiagrass pastures, was fertilized on June 10 with 30 kg N/ha 
using NH 4 NO 3 . On June 11, an area of 0.20 ha, to be used as grass- 



20 



21 



legume pastures, was seeded with aeschynomene at the rate of 9 kg/ha of 
unscarified seed. After aeschynomene seeding, grass -legume pastures 
were clipped weekly to an 8 -cm stubble until July 3 when the legume 
became established for successful association with the grass. Separate 
paddocks were fenced for the following grazing management by pasture 
type treatments: 

1) continuously-grazed bahiagrass pastures (0.08 ha) 

2) continuously-grazed bahiagrass -aeschynomene pastures (0.05 ha) 

3) rotationally- grazed bahiagrass pastures (0.05 ha) 

4) rotationally-grazed bahiagrass -aeschynomene pastures (0.05 ha) 

Each of these pasture type -grazing management combinations was 
replicated twice. 

Pasture management 

Pastures were managed to simulate continuous and rotational grazing 
systems. Continuous grazing was established by varying the stocking 
rate to maintain a 20-cm forage stubble height in the pastures at all 
times. Animals were removed from the continuously- grazed pastures when 
the target stubble height was reached and returned when the pasture 
height was about 25 cm. Rotational grazing was established by allowing 
a 4-week rest period followed by a grazing period that lasted until the 
forage stubble height reached 20 cm. Crossbred steers averaging 378 kg 
body weight were used to graze the paddocks as needed. Steers were kept 
in a bahiagrass pasture near the experimental site and were chosen at 



random when needed. 



22 



Sampling and Laboratory Procedures 

Pasture canopy and animal ingestive behavior measurements were 
obtained. Samples were collected for 4 cycles based on the grazing 
cycle of the rotational grazing system. A grazing cycle consisted of a 
4-week rest period and a 3- to 7-day grazing period. Each cycle 
corresponded approximately to the months of August, September, October 
and November for cycles 1, 2, 3, and 4, respectively (table 1). 

In rotationally-grazed pastures, canopy samples were obtained just 
prior to initiation of the grazing period (before rotational grazing, 
BRG) and just after conclusion of the grazing period (after rotational 
grazing, ARG) . Immediately after canopy measures were obtained, 
esophageally-fistulated steers were used for the collection of extrusa 
and measurements of ingestive behavior. Sampling this way permitted the 
investigation of animal defoliation effects during the grazing period in 
the rotationally-grazed pastures. Continuously-grazed pastures were 
sampled only once each cycle. Pasture canopy was sampled and animal 
ingestive behavior measured on dates similar to those when rotationally- 
grazed pastures were sampled (table 1). 

Forage sampling and separation 

A double sampling technique was implemented because uneven grazing 
was expected in these pastures. The double sampling technique included 
whole canopy visual estimations on 35 sites per pasture, of which 5 
sites were physically sampled. Visual estimations were made of herbage 
mass (HM, kg DM/ha) and percentages of grass, legume and other (weeds 
and dead) herbage. The 5 sites which were physically sampled were 



23 



Table 1. Sampling dates for canopy and ingestive behavior 
measurements 



Cycle 


Month 


Bahiagrass 




Bahia 


- aeschynomene 


CG a 


BRG b 


ARG C 


CG 


BRG 


ARG 


1 


Augus t 


11 


11 


16 


23 


18 


23 


2 


September 


22 


14 


18 


21 


21 


26 


3 


October 


20 


20 


22 


26 


26 


29 


4 


November 


21 


19 


23 


26 


25 


29 


Note : 


Numbers in table main body 


are days of the month 



a CG-continuous grazing 
b BRG-before rotational grazing 
c ARG=after rotational grazing 



24 



selected to cover the range in HM and percentages of grass and legume 
observed in the pastures. The samples were clipped in 10 cm layers. A 
half- square meter quadrat mounted in a frame was used. The quadrat 
could be adjusted vertically every 10 cm. Forage in the first 10 cm 
above ground level was not sampled. The maximum number of layers 
obtained from these pastures was 4; however, 3 or 4 layers in a pasture 
were not common. For statistical analysis by layer, the uneven number 
of layers in a pasture posed the problem of empty cells. In order to 
evaluate treatment effects between layers, layers 3 and 4 when present 
were pooled together with layer 2. Hereafter, the two layers used to 
describe the pasture canopy are referred to as bottom layer for layer 1, 
and grazing horizon for the pooled upper layers. The bottom layer was 
that between 10 and 20 cm above ground level and the grazing horizon was 
that over 20 cm above ground level. Layer data for cycle 1 were omitted 
from statistical analysis because, after pooling data for the grazing 
horizon, empty cells still existed. 

Clipped forage samples were stored at 0° C immediately after 
collection until hand separated. Samples were manually separated into 
grass leaf and stem, legume leaf and stem, dead material and weeds. If 
forage from a layer weighed 50 g or less for bahiagrass pastures or 75 g 
or less for bahia-aeschynomene pastures, forage in the entire layer was 
separated into its botanical components. When layer weights were 50 to 
100 g for bahiagrass pastures and 75 to 150 g for bahia-aeschynomene 
pastures, 50 and 75 g were separated, respectively. When layer weights 
were larger than 100 g for bahiagrass pastures or 150 g for bahia- 
aeschynomene pastures, one half was separated, but this was never more 



25 



than 100 g and 150 g for bahiagrass and bahia-aeschynomene pastures, 
respectively. 

For bahiagrass samples, leaf sheath and leaf blade were included in 
the leaf fraction. The bahiagrass stem fraction included only the 
reproductive stem and seeds when present. For aeschynomene the whole 
composite leaf was included in the leaf fraction. When legume pods and 
seeds were present, they were included in the stem fraction. 

Once separated, plant fractions were placed in a forced- air oven at 
60° C for 72 hr, then equilibrated to atmospheric moisture at room 
temperature, and weighed. In order to have enough sample for laboratory 
analyses, dry plant fractions were composited before grinding. Criteria 
for compositing was based on observations in the field at the time of 
sampling. Canopy layers were numbered from bottom to top. Two canopy 
layers from different sites were combined if they occupied the same 
horizontal space and if both sites were grazed or both were not grazed. 
If nongrazed, top layers were composited together independently of layer 
number. If the site was non-grazed and had only one layer, then forage 
in that layer was placed into the top layer composite, otherwise it was 
always included in the bottom layer composite. Intermediate layers were 
composited accordingly after selecting the top and bottom layers as 
described. Composites were then ground through a 1-mm screen in a Wiley 
mill. 

Dry matter (DM) , organic matter (OM) and crude protein (CP) 
analyses were conducted on the composited samples following AOAC (1975) 
procedures. In vitro OM digestion was determined by a modification 
(Moore and Mott, 1974) of the Tilley and Terry (1963) technique. 



26 



Correction of visually estimated data 

Actual whole canopy data were regressed on visual estimates to 
generate equations to correct the average of the 35 visual estimations. 
The best equations were selected based on the criteria of a low standard 
error of the mean and a high coefficient of determination. Plotting of 
residuals against predicted values, and the overlay of actual vs visual, 
and predicted vs visual plots were used to study the possibility of 
under- or over-estimation by the equation. Analysis of residuals and 
the studentized residual statistics were used as criteria for deleting 
outliers when needed. 

Correction of layer data 

Layer data were corrected before statistical analysis of the 
stratified pasture canopy was conducted. Because sites for canopy 
measurement were chosen to cover the range in canopy structure, layer 
data from these sites may not have been representative of the pasture. 

A formula relating adjusted and unadjusted whole canopy and layer values 
was used (Sollenberger et al., 1987a) to correct the layer data. This 
formula is based on the assumption that the ratio between an adjusted 
whole canopy (AWC) mean, which refers to the adjusted average of the 35 
visual estimations per pasture, and an unadjusted whole canopy (UWC) 
mean, generated from the average of the 5 actual clippings, is 
proportional to the ratio between an adjusted layer fraction (ALF, 
unknown) mean, and a corresponding unadjusted layer fraction (ULF) , 
which refers to the fraction layer values from the 5 actual clippings. 
Then, ALF is solved as follows: ALF - (AWC/UWC)*ULF. 



27 



Laboratory analysis values obtained from the composited layer 
samples described above were applied to the data in the following way. 
Dry matter and OM were applied to the air-dry weight raw data (ULF) . 
Crude protein and IVOMD values were applied to the OM, ALF weights. 

Then, from the total CP, in vitro digestible OM (IVDOM) and OM by layer, 
the percentage of adjusted layer CP and IVOMD were calculated. Whole 
canopy CP and IVOMD were calculated by summing the total CP and IVDOM by 
weight in all layers and obtaining the proportion of these relative to 
the total whole canopy OM. 

Ingestive behavior measurements 

Ingestive behavior measurements obtained included bites per minute 
(BPM), jaw movements per minute (JPM) , and intake per minute. Four 
esophageally- f istulated steers (average weight 410 kg) were used per 
pasture to obtain ingestive behavior measurements. Animals were kept on 
a bahiagrass pasture near the experimental site and were fasted 
overnight before being used for sampling. Sampling was carefully 
planned to avoid any biased pattern in the sampling rotation of animals 
between pastures. On the sampling day, cattle were allowed to graze the 
pasture for 15 minutes. During the sampling period, total bites were 
counted with a hand tally counter. A bite was defined as the sound of 
forage being severed at the moment of biting. Bites per minute were 
calculated from the total number of bites taken during the 15 -minute 
sampling period. Total jaw movements were counted with a digital 
microswitch counter which was connected to an elastic band placed around 
the animals muzzle. The opening and closing of the animal's mouth 
opened and closed an electric circuit which was recorded with a digital 



28 



device. Jaw movements per minute were calculated from the total jaw 
movements during the 15 -minute sampling period. From the BPM and JPM 
the number of actual bites per 100 jaw movements (B100JM) were 
calculated; this measurement was an indication of the degree of 
manipulation the animal exerted in order to obtain a bite of forage. 
Extrusa samples were collected during the sampling period. The diet 
sample was collected into a canvas bag fitted with a screened bottom to 
allow drainage of saliva. 

The total extrusa was weighed as collected, and a representative 
sub-sample of about 250 g was placed in a forced-air oven at 60° C for 
72 hr, then allowed to equilibrate to atmospheric moisture at room 
temperature, weighed and ground through a 1-mm screen in a Wiley mill. 
Laboratory DM, OM, CP and IVOMD were determined on extrusa samples as 
described above. Values for laboratory DM and OM were applied to the 
calculated total oven- dried extrusa weight. From the total number of 
bites and weight of extrusa collected during the sampling period, the 
bite weight (BW) was calculated. From BPM and BW, intake per minute 
(IPM) was calculated. 

Another representative sub -sample of about 250 g from the fresh 
collected extrusa was taken and immediately stored at 0° C. Diet 
botanical composition was determined from this second extrusa sub- 
sample. For this determination, a double sampling technique was used. 
Visual estimations were made by a microscope hit technique. For this, a 
10 to 15 g sample was washed gently with tap water in a sieve until the 
drain water was clear. The sample was then uniformly distributed in a 
16 x 45 x 5 cm tray which was about half filled with water. A 



29 



transparent plastic sheet with 50 hits (dot marks) was fitted in the 
tray on top of the water and sample. Hits were organized into 2 rows of 
5 (2 cm^) with 5 hits each (marks at 4 corners and center of a field). 
Plant parts under each of 50 hits were identified using a 6 . 3X 
stereoscope and entered into a computer as a numerical code. Plant 
parts identified were grass leaf and stem, legume leaf and stem, weeds 
and dead herbage. The same procedure used in the hand separation of 
forages, with reference to plant parts included in each botanical 
fraction, was followed for extrusa botanical composition. Samples were 
randomized for analysis in each of two runs. The second run did not 
begin until the first run for all samples was completed. Once the two 
runs were completed the codified hit data were processed by frequency, 
transpose and summary procedures in SAS-PC version 6 (SAS, 1987) for 
plant part percentage calculation. 

The mean of the two runs for each sample was calculated and the 
samples were separated in 10 percent-unit groups based on grass leaf 
percent. Twenty- five percent of the total number of samples were 
randomly selected, with similar number of samples from each group. 

These samples were completely hand separated for the same fractions as 
in the visual estimations . For this hand separation a 5 or 7 g extrusa 
sample was used from bahiagrass pastures or bahia-aeschynomene pastures, 
respectively. For this separation the same 6 . 3X stereoscope described 
above was used. Portions of the sample were placed in small aluminum 
pans which contained a small amount of water. The separated plant parts 
from a given sample were placed into aluminum pans and dried in a 
forced air oven at 105° C for 72 hr. Following this, samples were 



30 



placed into a desiccator for 2 hr and then weighed. Using the frequency 
procedure with the weight option and the transpose procedure of SAS-PC 
version 5.6 (SAS, 1987) the weight data were converted to percentages. 
The extrusa visual estimations were then corrected with the linear 
regression parameters obtained by regressing the hand separated values 
against the corresponding visual values. Selection of the best 
regression equation was based on the same criteria as mentioned above 
for canopy double sampling technique. This procedure was conducted 
separately for extrusa samples from bahiagrass and bahia-aeschynomene 
pastures; however, the correction for both type of extrusa samples was 
based on grass leaf percentage. 

Statistical Analysis 

Main effects for this experiment included pasture type (bahiagrass 
vs bahia-aeschynomene), grazing management (rotational vs continuous 
grazing; rotational included before and after grazing measurements to 
study the effect of animal defoliation) , canopy layer (bottom vs grazing 
horizon), and cycle (August vs September vs October vs November). The 
experiment was arranged in a split-split-plot design or in a split-plot 
design depending upon inclusion of canopy layer in the model (Snedecor 
and Cochran, 1980). The primary objective of taking repeated 
measurements over the season was to increase the accuracy of the study 
with a larger number of observations rather than to study seasonal 
effects. Most cycle interactions appeared to be related more to 
inconsistent variations between sampling periods than to a consistent 



31 



seasonal trend. So, only those cycle main effects or cycle interactions 
that had a biological meaning were discussed. 

For whole pasture canopy, diet composition and animal ingestive 
behavior data, pasture type and grazing management were included in the 
main plot with replicate nested within pasture type and with grazing 
management as the error term for main plot. Cycle was included as the 
subplot and the remainder as the error term. These data were analyzed 
using the General Linear Models procedure (GLM) of SAS-PC version 6 
(SAS, 1987). 

For pasture layer data, layer was included as the first sub-plot 
with replicate times layer nested within pasture type and grazing 
management as the error term. Cycle was included as the second sub -plot 
with remainder as the error term. The effects of interest from this 
analysis were those where layer was present. These data were analyzed 
using the analysis of variance procedure (ANOVA) in SAS-PC version 6 
(SAS, 1987). To further study the layer data analysis within layer was 
performed. Special interest in this analysis was when the between 
layers analysis indicated significant interactions. 

Contrasts for pair-wise comparisons and polynomial effects were 
performed. For means multiple comparisons, the Student-Newman-Keuls 
(Snedecor and Cochran, 1980) procedure at the 10% probability was used. 
Simple correlations were performed to study relationships between canopy 
characteristics and animal ingestive behavior. 



RESULTS AND DISCUSSION 



Pasture Canopy Structure and Forage Nutritive Value 
Whole Pasture Canopy 

Analyses of variance for whole canopy data are in tables 22 through 
24, appendix. 

Pasture type main effects 

Whole canopy herbage mass (HM) was greater in bahia-aeschynomene 
than in bahiagrass pastures (P-.006, table 2). Because canopy height 
did not differ (P-.762) between pasture types, the difference in HM was 
due to a more dense canopy in grass -legume pastures. Whole canopy 
forage IVOMD was greater in bahiagrass compared to bahia-aeschynomene 
pastures (P~.054; table 2). The average IVOMD of legume stem and grass 
leaf were 41.9 and 53.7% (not shown in table), respectively. 

Differences in whole canopy forage IVOMD between the two pasture types 
may have been due to the presence of legume stem, particularly in the 
bottom layer, which was less digestible than was bahiagrass leaf. 

Grazing management main effects 

Herbage mass was greater before rotational grazing than during 
continuous grazing (P<.10; table 3). The HM after rotational grazing, 
was not different from that during continuous grazing (P>.10). A 



32 



33 



Table 2 Least squares means for the effect of 

pasture type (bahiagrass (B) and bahia- 
aeschynomene (BA)) on whole canopy herbage 
mass, and in vitro organic matter 
digestion (n=24) 





Pasture 


Statistics* 3 


Item 3 


B BA 


SEM P 


HM, kg OM/ha 


969.4 1316.5 


59.14 .006 


IVOMD , % 


49.3 46.2 


0.91 .054 


3 HM-herbage mass ; 


; OM-organic matter; 


IVOMD-in vitro 



organic matter digestion 

k SEM=Standard error of mean; P-probability value 



34 



Table 3 Least squares means for the effect of grazing 
management (before (BRG) and after (ARG) 
rotational grazing, and continuous grazing) on 
whole canopy herbage mass, height and in vitro 



organic 


matter 


digestion 


(n-16) 






Item a 


Rotational 


Continuous 


Statistics^ 


BRG 


ARG 


SEM 


P 


HM, kg OM/ha 


1494. 0 C 


871. 5 d 


1063. 4 d 


72.43 


.002 


Height, cm 


28. 8 C 


17. 6 d 


18. 9 d 


0.62 


.001 


IVOMD, % 


50. 9 C 


47. 8 d 


44. 7 e 


1.11 


.021 



a HM-herbage mass; OM-organic matter; IVOMD-in vitro 
organic matter digestion 

b SEM-standard error of mean; P-probability value 
cde Means in a row with different letters are different 
(PC.IO) 



35 



similar trend was observed with respect to canopy height 
(table 3). The canopy was taller before rotational grazing than during 
continuous grazing (P<.10). As expected, there was no difference in 
canopy height when sampled after rotational grazing than during 
continuous grazing. Pastures under rotational and continuous grazing 
systems were grazed to a 20- cm stubble height. Although the actual 
canopy height obtained was slightly lower than planned it was still 
consistent with the objective of similar canopy height at the end of 
rotational grazing compared to that in continuously-grazed pastures. 

Whole canopy forage IVOMD of continuously- grazed pastures was lower 
than that of rotationally-grazed pastures ( P< . 10 ; table 3). Herbage 
mass, canopy height, and forage IVOMD were reduced as an effect of 
animal defoliation during rotational grazing (P<.10; table 3). 

Pasture type- grazing management interactions 

Canopy characteristics such as percentages of grass, legume, other 
herbage, and forage CP responded differently to grazing management and 
animal defoliation depending on pasture type. Grass percentage in the 
whole canopy of bahia-aeschynomene pastures was less than that in 
bahiagrass pastures due to the presence of legume in bahia-aeschynomene 
pastures (P-.001; table 4). Grass percentage in bahiagrass or bahia- 
aeschynomene pastures was lower in continuously- than in rotationally- 
grazed pastures (PC. 10). Grass percentage in rotationally-grazed 
bahiagrass pastures was reduced due to animal defoliation during the 
grazing period (PC. 10). In bahia-aeschynomene, however, this reduction 
did not occur. More legume was found in bahia-aeschynomene pastures 
before rotational grazing than when the pastures were grazed 



Table 4 Least squares means for the interaction of grazing management (before (BRG) and 
after (ARG) rotational grazing, and continuous grazing (CG)) and pasture type 



36 



o 

-4 

c 

cd 

u 

o 

X 

f*. 

a 

o 

c 

cd 

O 



o 

X 

9 

C 

o 



< 
c a 



y 

C 

y 

E oo 

O I 

c e 
x 

0 C 

01 i4 

<u 0) 
to u 
• o 

cd 44 

•H a 
X 

<d o> 
X -O 
3 

•d 44 

C o 
td 

•d 
~ C 
cq cd 



w C 
0) O 
Cd *r4 
44 4J 
bO *4 

cd oi 
t4 O 

x a, 

cd E 

jO O 
y 







rH 


H 


rH 


o 


w 


a 


O 


00 


VO 


oo 


o 




o 


o 


o 


O 


*r-i 




• 


. 


. 
















w 












*H 












4-> 




00 


CO 




O 


cd 


sc 


CO 


m 


oo 


<fr 


4J 


w 


• 


. 


• 




CO 


CO 


r—\ 


CM 


»— < 


o 










O 


T3 






00 




oo 


On 




o 


• 




• 


• 




u 




O 


T — 1 








co 


rH 


m 








o 




o 






o 


vO 


ON 


m 




< 




CM 


< — < 


m 




OQ 




**3- 












O 


o 


T3 


o 




o 




CM 


t— < 


CM 




oi 




• 


. 


. 




OQ 




CM 


CO 


o 








CM 


CO 


rH 






<D 




o 








r^ 




CO 


CO 




C3 


• 




. 


. 




u 


r — 1 




00 








m 
















T3 






C5 


on 




«— i 


CO 


cq 


3 


oo 




rH 








vO 




CO 








o 




d) 






BRG 


CM 




OO 


oo 




OO 




iH 


r>* 








<#> 


CM 








<#> 




<#> 








«. 


0 ) 








cd 


w 


g 


U 


<#> 




e 


w 




d) 






a) 


cd 


00 


rC 






■u 




0 ) 


•U 


Oh 




M 


o 


H-) 


o 


O 



O 

4J 

44 

<v 

4-1 

d) 

U 



o 0 

a> 



44 

o 

4-1 

CO 

y 

•H 

44 

CO 

•t4 

4-1 

cd 

■u 

00 



C <u 

-4 3 
y r4 
4-> cd 

0 > 
n 

a ^ 
w 

d) »r4 
T3 r— 4 

3 -4 
44 X 
y cd 

1 X 
C4 O 
U >4 

a 
•- l 
a a. 
cd 

•H • «. 
>4 C 
d> cd 

4J y 

cd E 
E 

44 
•O 
cd 

d) 

•d 

■O 

c 

cd 

01 
•d 



V 

a 



44 

c 

y 

44 

y 

44 

44 



y 

>4 

cd 

01 

u 

y 

44 

44 

y 



44 

c 

y 

44 

y 

44 



X 

44 



y 

a 

44 

y 

44 

3 

44 

CO 

cd 

a 

c 

44 *r4 

o x: 
y 44 

44 a 
44 S 



y 



3 

o 

44 

<d 



y "d 



O oo 

cd x 



44 C 
C "*4 
y 

E w 
y c 

bO cd 
cd y 
C 3d 
cd 

E y 
•d 
y 



37 



continuously (P<.10; table 4). Animal defoliation during rotational 
grazing reduced the proportion of legume in the canopy to levels found 
in continuously- grazed pastures (table 4). Other percentage, which 
included weeds and dead material, was greater in bahia-aeschynomene than 
in bahiagrass pastures when rotationally-grazed. No difference between 
pasture type for this botanical fraction was apparent when pastures were 
grazed continuously. Other percentage was greater under continuous than 
under rotational grazing management in bahiagrass pastures (P<.10). 

Other percentage in bahia-aeschynomene pastures, however, was greater 
under continuous than for the before rotational grazing sampling (P<.10) 
but similar to that after rotational grazing. An increase in other 
percentage was observed as an effect of animal defoliation in both 
pasture types (P<.10) indicating that animals were selecting against 
weeds and dead material. 

Bahia-aeschynomene pastures had higher forage CP than did 
bahiagrass pastures only before rotational grazing (Table 4), probably 
because of higher proportion of legume in the canopy at this time. 

Forage CP in bahiagrass pastures was not affected by grazing management 
or animal defoliation. 

In rotationally-grazed bahia-aeschynomene pastures forage CP was 
higher than that during continuous grazing only before rotational 
grazing (P<.10; table 4). The effect of grazing on CP reflected the 
reduction of legume due to animal defoliation during rotational grazing 
to levels similar to those in continuously-grazed pastures. Therefore, 
forage CP in bahia-aeschynomene pastures followed the response of legume 
to grazing management and animal defoliation. The levels of whole 



38 



pasture CP observed in all pastures, except for bahia-aeschynomene 
before rotational grazing, are on the border line of 7% suggested by 
Milford and Minson (1965) to be the critical level below which animals 
might experience depressed voluntary intake of dry matter due to protein 
deficiency. 

Cycle effects 

Legume percentage increased consistently during the season with a 
maximum in October and then dropped drastically in November (PC. 10, 
table 5). Rusland et al. (1988) observed in aeschynomene-limpograss 
pastures that maximum aeschynomene percentage occurred in October. The 
drastic decrease in legume percentage in November is probably related to 
flowering, seed development, and loss of leaf tissue. These results and 
other studies (Hodges et al., 1982; Sollenberger et al., 1987b) suggest 
that high quality forage is provided by aeschynomene from mid to late 
summer. 

Forage IVOMD decreased steadily across the season in both types of 
pasture (PC. 10; table 5). The decrease appeared to be at an increasing 
rate as the season progressed. Sollenberger et al. (1987b) observed a 
decline in forage IVOMD in bahiagrass pastures from June to October. A 
decrease in aeschynomene IVOMD as the plant matured has also been 
reported (Mislevy and Martin, 1985). 

Canopy Lavers 



Analyses of variance of canopy layer data are in the appendix: 
tables 25 through 28 for combined layer responses, tables 29 through 31 



39 



Table 5 Least squares means for the effect of cycle on whole 

canopy legume percentage in bahia-aeschynomene pastures 
(n-6) and in vitro organic matter digestion in all 
pastures (n-12) 







Cycle^ 






Statistics 0 


Item 3 


1 


2 


3 


4 


SEM 


P 


Legume , % 


8 . 6 f 


14. 0 e 


25. l d 


11. 7 e 


1.05 


.001 


IVOMD , % 


52. 5 d 


50. 0 e 


46. 6 f 


42.16 


0.88 


.001 



a IVOMD-in vitro organic matter digestion 
k Cycles correspond to August, September, October and 
November, respectively 

c SEM-standard error of mean; P-probability value 

def S Means in a row with different letters are different (PC. 10) 



40 



for within grazing horizon responses, and tables 32 through 34 for 
within bottom layer responses. 

Grazing management by layer interactions 

There was a grazing management by layer interaction for dead 
material percentage across both pasture types (P=.015; table 6). Dead 
material percentage was greater in the bottom layer than in the grazing 
horizon at the beginning of the grazing period in rotationally grazed 
pastures (P-.014). After rotational grazing and during continuous 
grazing, however, there were no differences in dead material percentage 
between layers. Dead material percentage was higher in both layers when 
the pastures were grazed continuously than when they were grazed 
rotationally (PC. 10). Dead material percentage increased in both layers 
as an effect of animal defoliation during rotational grazing (PC. 10). 

The increase in dead material percentage during rotational grazing was 
particularly high in the grazing horizon where there was a 100% 
increase as compared to only a 30% increase in the bottom layer. These 
results reflect the removal of green herbage by the animal particularly 
at the top of the canopy. The defoliation process is much easier at the 
top than at the bottom of the canopy where dead material may inhibit 
grazing. Therefore, it appears that grazing animals discriminate 
against dead material in the canopy and as the canopy is grazed it 
becomes even more difficult to obtain green material. Animals may spend 
more time searching for green herbage at the end than at the beginning 



41 



^ >1 

<3 (X 

04 O 

OQ C 

w (0 

o 

0) 

U T3 r-v 

O C C VO 
4-1 id O I 
cu C 

,o ^ *T3 v-> 

^ c 
O cd w 
U O CU 

C ^ ^ 4l 

CU CM 3 

5 b0r4 41 

a) C I w 

b0 14 C id 
« n ^ a 
C id 

id 4 w a 
B M# 

41 



bO 10 
C 3 

tH O 
N 3 
id C 
4l i4 
bO 4) 

c 

4-1 O 
O O ■ 
in cd 
C 3 si 4 



(X.C 

u 

r-4 10 
r4 CU 

cd cd 



G 

i-i cd 
4-) 

O 
cd 
4i 
CU 

U N Id 

C <d 4) 
-4 4l G 
bO <U 
(V o 
3H 4 
4> cd a) 



C 

4c O 
O 



O X 
4c cd 
O X 
- cd 
bO C 
C 4) i-l 
bO 

0) 
bO 
id 
4) 

C 
« 
o 



a 4t 

<u 



41 cd 
cd i-c 
U) 41 4l 
O (U 
4c 41 



c 

cd 



a) 




cd 






e 


/*N 


a 


T3 






o 




C 




co 


06 




cd 


-Q 


a) 


<i 


cd 




4c 






<u 


4-1 


0) 


cd 




T) 


cd 


>v 


3 


Vh 




a) 


cd 


cr 


a) 


C 


rH 


►J 


co 


iJ 


o 








<4-1 




a> 




■u 


cd 


>H 


p 




co 




0) 


3 




cd 






40 






c 


cd 


Q) 




hJ 


cd 


rH 


rH 





vo 

<u 

rH 

JQ 

cd 

H 



•U 

Oc 



4c 

0) 

►v 

3 

* 

s 



CO 



4t 

(U 

►v 

cd 

cJ 

* 

s 



41 

G 

cu 

e 

cu 

bO 

id 

C 

cd 

s 



u 

u 






o 

DC 

CQ 



id 

E 

ai 

4) 

























(0 






in 






rH 






00 










tH 






rH 






o 






CM 






CU 




CO 






o 






o 






o 






> 




>N 






• 






• 






• 






o 




r-H 
























-Q 




cd C 
























id 




c o 

Cd tH 






T— 1 


rH 




o 


VO 










B 




4J 






o 


o 




rH 


Ov 




ON 


ON 




o 




d) a 






o 


o 




o 


CO 




rH 


VO 








X cd 






• 


• 




• 


• 




O 


rH 




o 




4J >H 


















• 


• 




CM 




d) 




























M iJ 
























over 




o c 

<4H tH 






Csl 






CM 






CM 








CO d) 






ON 












oo 






G 




o .c 

•rl 






o 






o 






rH 






o 




•U 




rs 




















N 




CO U 




O 




















•tH 




•H O 




rH 




















4t 




iJ 4H 




• 


CO 


















O 




cd 




V 


m 




m 


rH 




oo 


o 




X 




4J CO 




fa 


CO 


in 




o 






00 


VO 








CO O 




'w' 


• 






• 






• 


• 




bO 




tH 






rH 


rH 




rH 


o 




rH 


rH 




C 




d) 4J 




4J 




















i-i 




CO 




c 




















N 




■U tH 




d) 


<D 


0) 
















cd 




4J 




}H 




<D 






4H 






u 




o cd 




d> 


rH 


o 


CM 


O 




o\ 


O 




On 


bO 




u u 




4-1 


• 


• 


O 


• 


• 


o 


• 


• 


00 






CO 




<4H 




On 


CO 


vO 


CM 


o 


CM 


00 


CM 


1 




}H 




tH 


CO 


CO 


* 






* 


rH 




• 


CM 




CU d) 
4 h „c 




T) 




















• - 




d> <u 




d) 




















T I 




u 




U 


4H 


4 h 




d) 






d) 






CU 

> 




o 

X 4J 




cd 


rH 


>d- 




00 


ON 






vO 


oo 


a> 








CO 


vO 






• 


• 


rH 


• 


• 


r^ 


r— 1 




Jh U 




in 


00 


CM 


CO 


O 


o 


ON 


CM 


O 






O d> 




(U 


CM 


CM 


• 








rH 


rH 




•o 




C4H 4H 




u 




















a) c 




d) 


4c 


4J 




















o 3 




CO }H 


(U 


d> 




















C O 




d) 


>v 


rH 


40 


40 
















0) 4t 




3 u 


cd 






T3 






T3 






4i bO 




rH d) 


»-4 


iJ 


O 


On 




oo 


CO 


rH 


VO 


VO 


CM 


(U 




Cd 




c 


CO 


• 


rH 


• 


• 


o 


• 


• 


rH 


4-i a) 




> cd 


>1 


d) 


rH 


o 


>d- 


CM 


o 


00 




O 


4-1 > 




X 


C4 


Jh 


rH 


CM 


• 


rH 






CM 


rH 




-r-l O 




•* -K 


O 


<u 




















•G X 




C S 


c 


4 h 




















cd 




cd 


cd 


4H 




















a> 




d> Jh 


o 


tH 




















x e 




e o 




-o 




















41 CJ 




4 h 


•o 


























4-1 


c 


X 


CM 






CM 












4-1 O 




O CO 


id 


4J 


rH 




rH 




CM 


rH 




O CM 




d) 


tH 




















i 






41 






















>vO 




O rH 


c 






















41 r- 4 




>H Cd 


(U 






















i-l 




LJ > 


E 


O 


OP 


















r-l 4l 




d> 


cu 


Lh 




















•r-l 0) 




• m 


bO >, 










OP 






OP 








T3 4i 


cd 4i 


cd 


cd 

■ 


















cd cs 


<u 


4i CU 


G *r4 














- 






X ft 


> 


<d S^v 


Id r-4 


C 


•rl 

i . 






4H 

cd 






a 






o 


CU 


■O cd 


E l4 


tH 


M 

a) 










d> 






4i E 


— 1 


C r-l 


X 








0) 






•u 






C4 O 




cd 


G cd 


CO 


cd 

6 






rH 






co 






1 41 


■v 


41 c 


cu X 


C 






d) 






d) 






[4, 41 


c 


0) i-l 


cu O 


<d 




fa 

fa 








fa 

fa 


[4| o 


3 


1 jC 


3 4i 


<u 


T) 

cd 

a> 

a 




So 

d) 

r-J 




fa 

fa 


40 

d> 

r-I 




M OQ 

a l 


o 

41 


s s 


4> o. as 

<u a 




HH 

O 

CL, 




M 

Q 

fa 




hH 

Q 

fa 


CL, r-l 

cd x 


bO 


CJ 


X Oc bO 
44 
•O cu 



42 



of the grazing period because they have to be more selective in order to 
obtain similar amounts of green material. The response of dead material 
percentage in the canopy layers to grazing management and animal 
defoliation is similar to the response of other material observed in the 
whole canopy of bahiagrass pastures (table 4). 

Legume leaf percentage was higher in the grazing horizon than in 
the bottom layer of bahia-aeschynomene pastures independent of grazing 
management and animal defoliation (P<.02; table 6). The difference 
between the layers was particularly high before rotational grazing. The 
percentage of legume leaf in the grazing horizon before rotational 
grazing was higher than that during continuous grazing (PC. 10). No 
difference in grazing horizon legume leaf percentage was found between 
after rotational grazing and continuous grazing. Legume leaf in the 
grazing horizon was reduced by animal defoliation during rotational 
grazing (PC. 10). No differences in legume leaf percentage were 
observed in the bottom layer due to grazing management and animal 
defoliation. These results on legume leaf percentage show the benefit 
of rotational ly- grazing bahia-aeschynomene pastures in terms of the 
accumulation of legume leaf at the top of the canopy, and suggest a high 
palatability of legume leaf by the animal. A similar observation was 
made by Sollenberger et al. (1987a), in aeschynomene- limpograss 
pastures . 

There was an interaction between grazing management and canopy 
layer on legume stem percentage (P-.028; table 6). Legume stem 
percentage was higher in the grazing horizon than in the bottom layer of 
rotationally-grazed bahia-aeschynomene pastures (PC. 10). There was no 



43 



difference, however, in legume stem percentage between the canopy layers 
of continuously grazed bahia-aeschynomene pastures. An effect of 
grazing management and animal defoliation on legume stem was found only 
in the grazing horizon (P— .019). Bahia-aeschynomene pastures under 
continuous grazing had a lower percentage of legume stem than did those 
under rotational grazing (PC. 10). Legume stem in the grazing horizon 
was reduced due to animal defoliation during rotational grazing (PC. 10). 
Higher consumption of legume stem from the grazing horizon as compared 
to the bottom layer might be related to differences in nutritive value 
of the legume stem. Legume stem in the top of the canopy appeared to be 
more tender and finer than that in the bottom layer and probably more 
palatable to the animal. Average IVOMD of legume stem in the bottom 
layer was lower than in the grazing horizon (38.1 and 45.6%, 
respectively; not shown in table). Sollenberger et al. (1987a) observed 
a high palatability of aeschynomene by animals grazing limpograss- 
aeschynomene pastures. Another possible reason for stem consumption 
might be that in grazing legume leaf the animal is forced to consume 
some legume stem. The ratio of legume leaf percentage to legume stem 
percentage in the grazing horizon was 0.52 and 0.20 for before and after 
rotational grazing, respectively, showing animal selection for legume 
leaf rather than for legume stem. Considering legume leaf and legume 
stem together the percentage of legume in the grazing horizon was 43.4, 
23.2, and 18% for before and after rotational grazing and continuous 
grazing, respectively. Comparing these values to the corresponding 
values for legume percentage in the whole canopy (table 4), the 
percentage of legume at the top of the canopy was about twice as high as 



44 



that in the whole canopy. Legume leaf /stem ratio decreased from grazing 
horizon to bottom layer, being 0.52 and 0.16, 0.20 and 0.10, and 0.50 
and 0.27 for before and after rotational grazing and continuous grazing, 
respectively. Similar observations were made by Moore and Sollenberger 
(1986) in limpograss-aeschynomene pastures. 

Pasture type, grazing management and canopy layer interactions 

Herbage mass was influenced by interactions between pasture type 
and grazing management (P=.038) and between pasture type and canopy 
layer (P-.005), but there was no three-way interaction (P-.877; table 
7). There was a type by management interaction in the grazing horizon 
(P— .005) but not in the bottom layer (P=.180). In bahiagrass pastures 
the grazing horizon had more herbage mass under continuous than under 
rotational grazing (P<.10), perhaps because of mature grass accumulation 
in areas rejected by animals in continuously- grazed bahiagrass pastures. 
In bahia-aeschynomene pastures, however, there was more herbage mass in 
the grazing horizon at the beginning of rotational grazing than in 
grazing horizon of continuously grazed pastures (PC. 10). This result is 
probably related to the high amounts of aeschynomene present at the top 
of the canopy at the beginning of rotational grazing. Herbage mass in 
the grazing horizon of bahiagrass and bahia-aeschynomene pastures was 
reduced by animal defoliation (PC. 10). 

The proportion of grass leaf in the canopy layers was influenced 
by grazing management, pasture type and canopy layer (P-.007; table 7). 

In bahiagrass pastures no difference in grass leaf percentage was found 



Least squares means for the interaction of pasture type, grazing management (before (BRG) 
and after (ARG) rotational grazing, and continuous grazing (CG)) and canopy layer on 
herbage mass and grass leaf percentage (n-6) 



45 







rJ 


















* 








r- 










X 








0 










* 


00 






O 








fa 


H 














o 


















w 




2 


m 


0 




<f 


m 




o 




* 


0 


00 




0 


0 




fa 




H 


0 






O 


0 




fa 






• 


• 






• 




t/5 


















fa 


















fa 




rJ 


cO 






Mt 






03 




•X 


O 






r-H 






fa 




2 


• 






• 






m 


X 


* 


r^ 














3 


H 
















cn 
























CM 




O 










2 


CM 


00 




vO 


00 








•X 


• 


• 






• 








H 




r^. 




CM 


r-H 










co 


















a> 








a) 










m 


CO 




On 












• 


• 


r^ 




• 


vO 






O 


O 


CO 


O 


CM 


m 


O 




a) 


CJ 




CO 


0 


CM 


co 


O 




C 




>d- 


0 












0 






fa 












6 


















0 


















C 




a> 








0) 










00 






O 


co 






x 


(3 


• 




r^- 




• 


OO 




a 


od 


0 


00 


O 




00 


0 


SJ 


w 




00 




O 


CM 


co 


0 




a) 






0 


• 










cd 






T— < 










fa 


1 
















e 


cd 
















a) 


•H 












T3 




6 


X 






r— 1 




00 






<D 


cd 


CJ 


• 


• 


m 


• 




«— i 


b0 


cQ 


fa 


t— 1 




0 


CM 


co 


0 


cd 




oQ 


»— 4 


vo 


0 


CM 




0 


C 






vO 


CM 










cd 








fa 










X 


















* 












fal 


fa 










VO 


O 




m 






/"■s 






• 




CO 




• 


00 


H 




CJ 


r— 4 




0 


O 




vO 






CJ 


CM 


m 


0 






O 








m 


00 


• 






. 


0) 


















u 


















3 


w 
















fa 


w 




fai 






a) 


a) 




w 


cd 




VO 


CO 




0 


r-H 




cd 


u 


CJ 


• 


• 


CM 




• 


O 


P-4 


b0 


fa 


in 




O 


m 


CO 


fa 




cd 


< 


CO 


O 


O 


m 


vO 


O 




fa 




CM 


vO 












rC 


















cd 


















OQ 






















<D 






T) 










CJ 


O 


VO 














fa 


• 


• 


CM 






O 






fa 


fa 


vO 


O 


av 


r— 1 


fa 












O 


V0 


r^ 


00 








co 




• 






. 






/ — - 




































' — 


















X 


















M 


















01 


CM 






CM 


t— 1 


























cfl 


















rJ 


























dP 












cd 


















rC 


















\ 






fal 












s 






cd 












0 






a) 


















r— J 












b£) 
















Cfl 






fa 


W 




fa 






E 






fa 


w 




fa 






01 


• 




M 


cd 




fa 






4-1 


s 




Q 


u 




Q 






t— C 


X 




fa 


O 




fa 



3) 

> 0 ) 
oi X 

r-i m 

•o u 
c o 

3 4-1 
O 

M Cfl 
00 O 
fa 

4) 4J 

> cn 
O -M 
X 4-> 
cfl cO 

4-1 
E to 
a) o 

0 ai 
C o X 

(0 CM 4-> 
U 

01 m o 

CM 01 4-1 

> 
o 



Cfl 

o 

•M 

4-1 

0) 

M-C 

4-1 

cfl 

4-1 

CO 

<U 

X 



o 

4-1 



M tM 
<U U ■ 
<4-1 (U V 
CU 

M CO 



cm 



cu 

X 



M 

<U 

C <M 
o ai 
_ N U 
4-1 -M 

, U S 

<M O * 

O X H 

MiO 4 
4-1 C O 
•m -m cm 

r-4 N 
•M CO W 
XI U CU 

CO 00 3 
X I r-4 

0 CM CO 

U > 

Ou •- 

1 r-4 -- 
fa CU 

fa > 4J 
H 41 *4 
Q T— t I— f 
fa -M 

■O X 
•- C CO 
U 3 XI 
a) O o 

4-1 M M 

oo a 

I 

ai fa 
> 

o •- 
fa xi c 

C cd cfl 

cfl ai 

00 B B 
U o 

0 

1 o 

2 CM 

o ■ 
o 



X 

* 

X 

* 

H 

M 

0 
CM 

Cfl 

01 
3 

i — i 

cfl 

> 



4-1 

c 

a <u 
o u 
C 01 
co cm 
o cm 

•H 
■O T> 

c 

' 01 
M 
<0 

Cfl 
M 
01 
4-1 



4-1 

c 
01 
S 
01 
00 4-1 
cfl 01 
C fa 
cfl 

S 4-1 

e 

00 01 
c u 

M 01 
N CM 
CM 



cfl 

00 M 
C O0TJ 

•M 

N - X 
CO 01 4-1 
U fa fa 
00 ^ 3 
4-1 



01 

a 

>■> 

4J 



4-1 

CO 

E 



C oi 
cfl M 
3 

01 4-1 

cfl ai 

>•. cfl 
4J 



cm 

O 



fa 3 
4-1 
C cfl 
ai co 
cu Q, 

« C 
01 



0.-0 X 

4-1 

C C fa 
oi o 3 
01 



cfl M 
Cfl M 01 
CO 0) 

E 

cO M 

0) fa 

00 

CO E 
XI O 
M 4J 

01 4-1 

x o 
JL x 



3 4-1 
4-1 CJ 
0) cfl 
X n 



3 

o 

u 



cfl 

■o 

c 

Cfl 

4-1 

Cfl 

I 



01 

4-1 

c 



X fa co 
cfl X O 



4-1 

0 ai to 
Cfl X C 

M 4J cfl 

01 01 
M M S 
C O 

-m cm cm 
01 
•3 



46 



between layers at the beginning of rotational grazing. After rotational 
grazing (P=.010) and under continuous grazing (P=.068), however, the 
bottom layer had higher grass leaf percentage than did the grazing 
horizon, indicating a more intense defoliation of upper layers by the 
animals. In bahia-aeschynomene pastures the percentage of grass leaf in 
the bottom layer was higher at all times than that in the grazing 
horizon (PC. 01), no doubt reflecting the presence of aeschynomene in the 
top of the canopy. 

In both layers of bahiagrass pastures, continuously- grazed pastures 
had lower grass leaf percentage than did rotationally- grazed pastures 
(PC. 10; table 7). In bahia-aeschynomene pastures a difference between 
the two grazing systems occurred only in the bottom layer with lower 
grass leaf percentage under continuous grazing than before rotational 
grazing (PC. 10). No difference in grass leaf percentage was observed 
between the continuous and after rotational grazing in the bottom layer 
of bahia-aeschynomene pastures. Grass leaf percentage in both layers of 
bahiagrass pastures was reduced by animal defoliation (PC. 10). A 
similar effect occurred only in the bottom layer of bahia-aeschynomene 
pastures. The proportion of grass leaf in both layers of bahia- 
aeschynomene pastures appeared to be lower than that in similar layers 
of bahiagrass pastures (table 7) because of the presence of legume in 
bahia-aeschynomene pastures. 

Forage CP and IVOMD were affected by interactions of grazing 
management, pasture type and canopy layer (P=.016 and .079, 
respectively; table 8). In bahiagrass pastures at the beginning of 



47 









rJ 




















* 


VO 






ON 






✓“s 






as 


ft 






r** 






CJ 






* 


o 






O 






Oi 




a 


H 


• 












CQ 




















- — - 


O 


















c 


w 




3S 


CO 


oo 






CM 




4) O 


o 




* 


vO 


ON 




vo 


oo 




14 


•ft 




H 


O 


CO 






CM 




O 14 


V 






. 


. 






. 




44 41 


w 


















41 


*H 


















X> cO 


U 




a 


CO 






o 






^ r-l 


cd 




* 


CM 






vO 








4-> 




S 


• 












4-> !>v 


CO 


y 


* 


O 






O 






c a 




taj 


H 














41 O 




CO 
















s c 




















41 cd 








o 


oo 




<t 


O 




bO O 






a 


uo 


CO 




co 


00 




cO 






* 


• 


• 




. 


• 




C T3 






H 


o 


o 




ft 


ft 




cd C 




















E cd 




















bO rN 








0) 






d) 






C — 








uo 


o 




r^ 


ft 




*rl O 












in 


• 




o 


N C_> 






O 


on 


00 


co 


CM 


CO 


in 


CO '-r' 




a) 


U 






O 






vo 


14 




$2 








*. 






• 


bO bO 




a> 
















C 




S 
















- .rl 




o 
















4) N 




c 




CD 






d> 






a co 








in 


m 




CO 


O 




14 






O 


• 


. 


ON 


• 




r^ 


4-1 bO 




o 


a 


ON 




ft 




CO 


o 




£ 


CO 








o 






CM 


4) 01 




0) 








. 






• 


14 3 




cd 
















3 O 


4J 


i 
















4J 3 


c 


cd 
















oi C 


0) 


•H 




T> 






T3 






cd *rl 


B 


42 




vO 


vo 




O 






a 4-i 


d> 


cd 


CJ 


• 


• 


ft 


. 


• 


00 


G 


bO 


CQ 


oi 


CO 


00 


o 


o 


VO 


CM 


44 O 


co 




« 


ft 




o 


m 




O 


o o 


C 










. 










cd 


















C TJ 


SS 


















O C 




















•rl CO 


* 






0) 






d) 


d) 




4-1 








o 


m 






CM 




o - 


/^s 








• 




« 


. 


>d* 


Cd bO 


H 




o 


r-» 


VO 


>d- 


CM 


CO 


CO 


14 C 






u 






ft 






vo 


41 *rl 












. 






• 


4-1 N 


d) 


















C <0 


J-t 


















•H 14 


3 


CO 
















bO r 




co 




a) 






<D 


T3 




41 VO 


co 


cd 




vO 


CM 










JZ r-l N 


cd 


u 


o 


• 




00 






ft 


*J cd C 




bO 


a 




r^ 


CO 


vo 


ON 




C V— ' 




cd 


•3 






ft 




vf 


o 


14 O 




•ft 








. 








O -rl 41 




42 
















44 4J 3 




cd 
















CO r4 




CQ 
















01 4-1 Cfl 








T) 








72 




C O > 






o 


on 


vo 




m 


VO 




CO 14 






a 


• 


• 


o 






CM 


41 41 






a 


00 




ft 


co 


o 




E rv > 












O 


m 


m 


o 


O -rl 












. 








01 Oi 4-1 






r-\ 














41 < -rl 






a 














14 v-' 14 




















CO 4J 




















3 14 3 






a 














cr 4i c 






14 














01 4-1 






41 


CM 


»—t 




CM 


ft 




44 41 






>-, 














4J Cfl bO 






cfl 














01 cfl 






a 














cfl 3 H 




















4) C O 




















r4 Cfl 44 


































c#P 






CXI 




















41 






cfl 


dP 




Pn 


Q 






H 






S 








S 




Ui 


rQ 






41 


•* 




M 


O 




M 


dj 






4-1 






Q 


> 




Q 


H 






M 


o 




Oi 


M 




Qh 



c 

o 

V 

0] 

a) 



<u 
> 
bo a) 



01 

> 

ft 

(0 

c 

<d 

a> 

JZ 

4J 

n 

o 

44 



•3 

§ 

o 

14 

bO O 



C JZ 
CO c0 
bO 

u B 
o o 



o 

u 

4-> 



Q 

s 

o 

> 



4-1 

01 

•rl 

4-1 

<0 

4J 

01 

4) 

SZ 

4-1 

O 

4-1 

i-i 

4) 

4-1 

4) 

1-1 

as 

5 

H 



4) bO 4 



U 

c 



c o 

4-1 

4) N 

1-1 4) 01 

4) l-l 41 

4-1 bO 3 

44 | r-l 

•r4 <N CO 
X) > 



41 

JZ 

•u 



u 



41 ►, 
> W 

41 *r4 



O -4 
T3 £> 
^ C cfl 
" 3 J3 
O O 
l-l 14 

bo a 

i 

41 “ 

> 

O 
-Q 
cO 



cO 

£> 

0 
u 
a 

1 

a 

a 



Q O 
04 CM 



•- O 



c 

cO 

41 

E 

44 

O 

14 

O 

14 

14 

4) 



41 14 
4J v 

o ^ -o 

14 CO 14 

a. 



4) 

T3 

3 

14 

0 

1 

a 

o 



cfl 

TO 

C 

cO 

4-1 

01 



on 
CO O 



41 

■o 



within layer; Values for T*M*L refer to the statistics for the interaction between pasture type, 
grazing management and canopy layer 

Means in a row within pasture type with different letters are different (P<.10) 



48 



rotational grazing, forage CP in the grazing horizon was higher than 
that in the bottom layer (P-.010). No differences in forage CP between 
layers of bahiagrass pastures were observed, however, at the end of 
rotational grazing or under continuous grazing. In bahia-aeschynomene 
pastures, the grazing horizon always showed a higher forage CP than did 
the bottom layer (PC. 04). In the grazing horizon of bahiagrass pastures 
forage CP was higher before rotational grazing than under continuous 
grazing (PC. 10). No differences in CP, however, were found in the 
grazing horizon between after rotational grazing and continuous grazing. 
Forage CP in the grazing horizon of bahiagrass pastures was reduced by 
animal defoliation (PC. 10). No differences were found in forage CP in 
the bottom layer of bahiagrass pastures due to grazing management or 
animal defoliation. 

In bahia-aeschynomene pastures, forage CP in the grazing horizon 
was lower under continuous grazing than before rotational grazing 
(PC. 10); table 8). No difference in forage CP was found, however, for 
grazing horizon between continuous grazing and after rotational grazing. 
In these pastures, like in bahiagrass pastures, the bottom layer did not 
differ in forage CP whether the pastures were continuously or 
rotationally grazed. Animal defoliation reduced forage CP in the 
grazing horizon of bahia-aeschynomene as was observed in bahiagrass 
pastures. These responses to grazing management indicate a benefit in 
terms of forage CP in the top of the canopy when bahiagrass or bahia- 
aeschynomene pastures are grazed rotationally. This offers the 
opportunity of a higher forage CP for animals at the beginning of the 
rotational grazing period. The benefit of overseeding aeschynomene 



49 



inbahiagrass pastures is also shown where forage CP of the grazing 
horizon is maintained high independent of grazing management. Of 
course, the levels of forage CP will depend on the proportion of legume 
in the canopy. The advantages of rotational grazing and overseeding of 
aeschynomene in terms of forage CP might not be readily obvious when 
looking at the whole canopy. 

Forage IVOMD in the grazing horizon was higher than that in the 
bottom layer of bahiagrass pastures before rotational grazing (P-.042; 
table 8). After rotational grazing forage IVOMD in the bottom layer of 
bahiagrass pastures was higher than that in grazing horizon (P-.041). 
Lower IVOMD in grazing horizon after grazing may be due to defoliation 
of the grazing horizon which did not occur in the bottom layer. Canopy 
layers in continuously- grazed bahiagrass pastures did not differ in 
forage IVOMD but IVOMD was lower than when rotationally- grazed. 

In bahia- aeschynomene pastures, the grazing horizon had higher 
forage IVOMD than did the bottom layer before rotational grazing 
(P-.028 ; table 8). No differences in IVOMD were found between layers, 
however, either after rotational grazing or when pastures were grazed 
continuously. Forage IVOMD in the grazing horizon of bahia- 
aeschynomene pastures before rotational grazing was higher than that 
when these pastures were grazed continuously (P<.10). Animal 
defoliation during rotational grazing reduced forage IVOMD in the 
grazing horizon to levels similar to those found in the continuously- 
grazed bahia- aeschynomene pastures (PC.10). Forage IVOMD in the bottom 
layer of bahia -aeschynomene pastures was not affected by grazing 
management or animal defoliation. Forage IVOMD appeared to be somewhat 



50 



lower in bahia-aeschynomene pastures compared to bahiagrass pastures, 
particularly in the bottom layer of rotationally- grazed pastures, 
probably because of the presence of less-digestible legume stem. The 
responses of forage IVOMD to grazing management indicate, again, a 
benefit of rotational grazing for bahia-aeschynomene and, particularly, 
bahiagrass pastures. 

Cycle effects 

There was an interaction between canopy layers and cycle on legume 
leaf percentage (P-.002) and legume stem percentage (P-.004) in bahia- 
aeschynomene pastures (table 9). In September and October legume leaf 
percentage in the grazing horizon was higher than that in the bottom 
layer (P-.001). In November, however, no difference in legume leaf 
percentage was found between canopy layers. No difference in legume 
stem between layers was found in September. In October and November, 
however, the percentage of legume stem in the grazing horizon was higher 
than that in the bottom layer (PC. 01). The data illustrate stages of 
growth of the legume across the season. Early in the season the legume 
plants are young and small with a high leaf/stem ratio. As the plant 
matures, larger absolute amounts of leaf and stem are found in the 
pasture; however, the leaf/stem ratio declines. 

Legume leaf percentage and stem percentage in both layers of the 
canopy changed with cycle (PC. 10; table 9). In the grazing horizon, 
legume leaf percentage was maximum in September and October, then 
declined in November. Legume stem percentage in the grazing horizon and 
legume leaf percentage and legume stem percentage in the bottom layer 
increased with maximum levels in October and then declined in November. 



Least squares means for the interaction of cycle and canopy layer on 
legume leaf and stem percentage in bahia-aeschynomene pastures (n— 6) , 
and on dead material percentage across all pastures (n-12) 



51 





Vi 






















4) 
























CN 












m 








<d 


o 






O 






CO 








H-] 


o 






O 






O 








* 


• 






• 












•d 


u 




















CL, 


























CNJ 


On 




CNJ 


rH 




rH 


rH 






c_> 


o 


O 




o 


o 




o 


o 








o 


O 




o 


o 




o 


o 






u 






















4) 


o 












r- 










On 




















<d 


• 






• 












o 


rJ 


o 






rH 






rH 






E 


* 




















w 


u 




















cn 


























o 


"d- 




00 


^d- 




CO 


o 








CN 






CO 


NO 




NO 


CO 






u 




• 




















rH 


o 




CNJ 


o 




rH 


rH 








<4-1 


00 




<4H 


<4H 




a) 


a) 








oo 


'•d’ 


r>. 


rH 


CO 


CN 


00 


m 


O 






• 




ON 


• 


• 


O 


• 


. 








rH 


o 


CNJ 


OV 


ON 


O 


00 


00 


00 










* 


rH 




• 


CO 


CO 


• 


CN 






















o 






















S-C 




0) 


a) 




<D 


a) 




<4H 


<4H 








on 


vf 


rH 




o 


rH 


rH 


oo 


m 


o 






• 


o 


• 


• 


O 


. 




rH 


<D 


CO 


rH 


CO 


o 


rH 




o 


rH 


00 


o 


r— i 
M 




rH 




• 


CO 


rH 




CN 


CN 




w 






















o 


























<D 


<4-1 




00 


<4H 




00 


oo 








on 


00 


rH 


m 




ON 


CO 


o 


rH 




CM 


• 


• 


o 


• 


• 


CO 






o 






o 


rH 


o 


ON 


ON 


ON 


VO 


CN 


O 






rH 




* 






• 


rH 


CN 












































<D 






















CNJ 


rH 




CN 


rH 




CN 


rH 
























kJ 


































dP 








<#> 






dP 


























iH 








- 






- 






cd 








<4-1 






E 






•H 








cd 






0) 






u 








<D 






4J 






a) 








r— 4 






W 






■u 




















cd 








<D 






CD 






6 






cd 


E 




Ch 


£ 




Ph 






pL, 


6 


n 




Uh 


3 




Ph 


T) 




Ph 


0) 


00 




h 


oo 




H 


cd 




M 


-U 


a) 




Q 


a) 




Q 


a) 




Q 


M 






D-, 


H-J 




CL, 


Q 




CL, 



a 

o 

o 

CM 

Vl 

<B 

> 

O 

c 

o 

N 

•H 

vi 

o 

X 

bO 

c 

•H 

N 

cd 

Vl 

40 

CM 



4) 

> 

a> 



■o 
a) C 
o d 
C o 
a> u 
Vi m 
0) 

4-1 4) 
4-1 > 
•r4 o 
X) X 



4) 

X 

4-1 



cd 

B 

o 



4-1 O 
O CM 
• rH 
f^O 4) 
4-) i-\ > 
•r4 4) 

rH Vl i — I 

•r4 44 

X 

cd cd C 

0 

u a 

a. O 

1 u 

h 44 111 

o > 

M CQ O 
Q I X 
cu <-h qj 

cd 



d 

o 

V 

40 



4H 

•O 4) 



Cycles correspond to September, October and November respectively 
SEM-standard error of mean; P-probability ; Values for C refer to the 
statistics for the analysis within layer; Values for C*Layer refer to the 
interaction between cyle and canopy layer 

6 Means in a row with different letters are different (PC. 10) 



52 



The October increase is most evident for legume stem in the grazing 
horizon, and is similar to whole canopy legume percentage which reached 
maximum levels in October (table 4). Similar observations were made by 
Rusland et al. (1988). Legume stem increased by about 300% in the 
grazing horizon and by about 75% in the bottom layer from September to 
October. The decline of the legume in November was due to a decline of 
legume leaf percent in the grazing horizon of about 600% and to a 
decline of legume stem percent of 60% in the grazing horizon and of 80% 
in the bottom layer. The decline in legume in November is due to the 
reproductive stage of the plant. During this month the partition of 
photosynthate is probably directed towards the development of 
reproductive organs and seed production, and there is loss of leaves. 

There was an interaction between canopy layers and cycle for dead 
material percentage across all pastures (P-.035; table 9). Dead 
material was higher in the bottom layer than in the grazing horizon in 
September and October (P<.02). No difference in this respect was found, 
however, in November. Accumulation of dead herbage at the end of the 
season illustrates the decline in the rate of growth of bahiagrass with 
the advance of the season and also the decrease in legume percentage in 
November. 

There was an interaction between cycle, pasture type and canopy 
layer for weed percentage (P— .009; table 10). In bahiagrass pastures 
there was no difference between layers in September but there was 
thereafter with an increasing difference up to November (P<.04). Weed 
percentage increased in the grazing horizon (P<.10) with no change in 
the bottom layer. The grazing horizon had higher percentage of weeds 



Table 10 Least squares means for the interaction of cycle, pasture type, 
and canopy layer on weed percentage (n— 12) 



53 



























CU 








* 


ON 




















X 








H 


o 
















a 




P C 








* 


o 
















o 




(U 








u 


• 




















P <U 






Q) 
















a) 




o 




O 3 






CU 
















C 




CNI 




4-1 P 








H 


f-M 


00 










a) 








(U 








* 


o 


CO 










6 




P 




01 X 








CJ 


o 


m 










O 




<u 




U 










• 


• 










C 




> 




r— C »r4 C 






















>N 




o 




<U P O 






















X 








> 91 *r4 








1-3 


vO 












o 




c 




•r-C -P P 




/-N 




* 


o 












C/3 




o 




U il u 




o 




H 














a> 




N 




O CO co 




rM 




* 


CM 












cd 




•H 




CU P P 






'd 


o 














i 




p 




Q. W CU 




V 


X 
















cd 




o 




01 P 




IX, 


w 
















•rM 




X 1 




<U <U 0 




V-/ 


C/3 




oo 


oo 










x: 








P X -p 








H 


co 


VO 










cd 




bO 




p 




4J 




* 














X) 




c 




P 0) 




C 




o 


CM 


r — 1 










H 




•H 




CU O X 




a> 


















• 




N 




X P P 




Mi 


















>o 




cfl 




s 




a> 


















X 




P 




(U P O 




Mm 






Mm 












a 




bO 




> CU P 




Mm 






CM 


o 


r— 1 


O 


VO 


co 


w 




y 




O 4-1 




•H 






• 


• 


o 


• 






a) 




CM 




Z <U P 




T3 






m 


CO 


o 


00 


in 




cd 








P CU 










CM 


i-M 


• 


r-M 


r-M 




i 




• - 




13 4-1 




a> 


















cd 




r4 




C H <U 




Mi 


















•rH 




0) 




cO * P 




cd 


CJ 
















X 




> 




O 






v — ' 
















cd 




<u 




P kJ 




w 






00 












PQ 




r-4 




CU P * 




Ml 


o 




co 


t— 1 


vO 




VO 












-Q O H 


Mi 


a) 


0) 




• 


• 


CO 


• 




CM 


• «. 




•a 




O 44 -it 


<u 


M> 


I-M 


co 


00 


CM 


O 


o 


vO 


CM 


0) 




c 




P U 


>> 


4J 


o 




r-M 


r-M 




CM 


T”M 


• 


o 




3 




O c/l 


cd 


a) 


>5 
















c 




o 




O <U P 


i-M 


i-M 


CJ 
















a) 




p 




3 O 


























bO 




- r-4 44 




U 


















a) 








P CO 


CM 


C 


















MM 




4) 




<U > CO 


o 


Q) 






On 




VO 


l"M 


ON 




tMI 




> 




XI <U 


c 


u 




CM 


• 


• 




• 


• 


r-M 


•tM 




O 




E •- 3 


cd 


a) 






00 


ON 


00 


m 


in 


o 


T3 




43 




CU C 1-4 


o 


Mm 










• 


CM 


T“M 


• 






cO 




P CO CO 




Mm 


















a) 








a cu > 


T3 


•H 






















E 




cu a 


C 






►J 














u 




o 




CO 


cd 




























44 P 




X 


















MM 




o 




O O CU 


» 


■u 




x 














O 




CM 




P 


a) 03 


*tM 




u 


















• 


r-M 


P CO 


ex 3 






0) 


CM 


r— 1 




CM 


l“M 




>o 




o 


<u 


13 O rP 






















4J 




T— ) 


> 


C P 


M» cd 






cd 














•tM 






0) 


O P C 


> 


o 




X 














r-M 




p 


i-M 


O. <U -P 


<D 


Ml 


















•H 




0) 




91 X 


Ml 




















X 




T) 


OOP 


d 4J 


cd 












• 






cd 




cd 


c 


P P -4 


4J iM 














>0 






X 




r— i 


3 


P co 3 


W i-M 


C 












X 






o 






o 


O 13 


Cd *iM 


•*M 






03 






o 






Mi 


c/3 


e 


p 


O C co 


ex X3 






/•"N 


0) 






w 






a 


<u 


o 


bO 


CO 14 


cd 


C/3 




H 


cd 






0) 






i 


Mi 


p 




91 P C/l 


- £1 


c 


Q) 


'w' 


Ml 






cd 






Cm 


d 


p 


0) 


<U 91 


(D o 


cd 


u 




00 






i 






Cm 


■p 


o 


> 


r4 1 r-4 


r-l Mi 


<u 


3 


cd 


cd 




Cu 


cd 




Pm 


hM 


C/3 


X 


o 


O S CO 


u as 


■U 


0) 


tM 




U* 


•h 




Pt| 


Q 


cd 




X 


>mW c 


N 




c/3 

cd 


ex 

>5 


X 

cd 




M 

Q 


cd 




M 

Q 


CM 


a.-4 


tO 


U 9) CO 


a oix: 
bO 


PM 


X) 


CQ 




CU 


0Q 




CM 


cd 




X 




O 13 


CU 44 



54 



than did the bottom layer in bahia-aeschynomene pastures in September 
(P-.014). No differences in this respect were found thereafter. 

Diet Botanical Composition and Nutritive Value 

Analyses of variance for diet data are in tables 35 through 37, 
appendix. 

Pasture type main effects 

A higher percentage of weeds was found in the diet of animals 
grazing bahia-aeschynomene than in those grazing bahiagrass (P-.012; 
table 11) . The weed percentages in the grazing horizon of bahia- 
aeschynomene pastures and bahiagrass pastures were 21.3 and 17.5%, 
respectively (P-.144). The data suggest that animals selected against 
weeds . 

Dead material percentage was higher in the diet of animals grazing 
bahiagrass pastures (P-.054; table 11). The canopy data did not 
indicate a higher dead material percentage in bahiagrass as compared to 
bahia-aeschynomene pastures (P-.215) and the means in the grazing 
horizon were 24.5 and 26.3%, respectively. This suggests that animals 
did not discriminate against dead material in bahiagrass pastures as 
well as they did in bahia-aeschynomene pastures. Most dead material in 
bahiagrass or bahia-aeschynomene pastures was of bahiagrass origin. So, 
dead material is well mixed with green grass making it more difficult 
for the animal to select only for green material. 

Animals grazing bahia-aeschynomene pasture consumed a diet that was 
greater in CP percentage than did those grazing bahiagrass (P-.003; 
table 11). The difference in diet CP was likely due to the presence of 
legume in the bahia-aeschynomene pastures. The calculated forage CP 



55 



Table 11 Least squares means for the effect of 

pasture type (bahiagrass (B) and bahia- 
aeschynomene (BA)) on botanical 
composition and nutritive value of the 





diet ingested by 


animals 


(n-24) 






Pasture 


Statistics* 5 


Item a 


B 


BA 


SEM 


P 


Weeds, % 


7.9 


13.0 


1.01 


.012 


Dead material, % 12.2 


9.2 


0.91 


.054 


CP, % 


9.5 


13.9 


0.61 


.003 


IVOMD, % 


56.6 


52.3 


0.61 


,003 



a CP— crude protein; IVOMD— in vitro organic matter 
digestion 

k SEM-standard error of mean, the highest value is 
shown; P-probability value 



56 



average in the whole canopy (table 4) was 7.5 and 8.6% for bahiagrass 
and bahia-aeschynomene pastures, respectively. Animals were able to 
select a diet higher in CP than that on offer in the whole canopy 
(table 4) and in the grazing horizon (table 8). The differences in 
forage CP between pastures, however, were reflected in differences in 
diet CP. 

The IVOMD of the diet consumed by animals grazing bahiagrass was 
higher than that consumed by animals grazing bahia-aeschynomene pastures 
(P-.003; table 11). The difference in diet IVOMD was similar to that 
observed for whole canopy forage IVOMD (table 2). According to canopy 
layer data (table 8) the difference in forage IVOMD between pasture 
types was in the bottom layer rather than in the grazing horizon. The 
difference in the bottom layer, however, was not consistent. The 
presence of legume stem in bahia-aeschynomene pastures, particularly in 
the bottom layer, might be related to any forage IVOMD difference 
between pasture types. 

It is important to note that animals were able to consume diets 
with higher CP and IVOMD percentage than those of the whole canopy 
(table 2 and 4) and canopy layers (table 8). These data indicate animal 
selection of the more nutritious botanical components in the pasture. 
Grazing management main effects 

Animals on continuously- grazed pastures consumed a diet greater in 
weed percentage than did those on rotationally-grazed pastures (P-.065; 
table 12). The difference appears to be similar to that observed in 
the whole canopy where continuously- grazed pastures had higher 
percentages of other materials than did rotationally-grazed pastures 



57 



Table 12 Least squares means for the effect of grazing 
management (before (BRG) and after (ARG) 
rotational grazing, and continuous grazing (CG)) 
on diet botanical composition and nutritive 
value (n-16) 



Item a 




Management 


Statistics^ 


BRG 


ARG 


CG 


SEM 


P 


Weeds , % 


8 . 6 d 


9 . 2 d 


13. 4 C 


1.24 


.065 


Dead material, % 


5 . 6 d 


14. 3 C 


12. 3 C 


1.11 


.003 


CP, % 


14. 2 C 


9 . 6 d 


11.3 d 


0.75 


.015 


IVOMD, % 


55. 6 C 


55. l c 


52. 4 d 


0.75 


.048 



a CP— crude protein; IVOMD— in vitro organic matter 
digestion 

b SEM-standard error of mean; P-probability value 
cd Means in a row with different letter are different 
(P<.10) 



58 



(table 4). The layer data indicated, however, that the difference in 
other percentage was related to differences in dead material rather than 
weeds (table 6). An observation from the field was that animals on 
continuously-grazed pastures grazed mostly from new regrowth in short 
canopy areas that had been grazed before. In these shorter canopy areas 
the percentage of weeds appeared to be relatively high as compared to 
rejected spots of more mature forage in the pasture. Grazing short 
spots might be the reason for the higher percentage of weeds in the diet 
of animals under continuous grazing. The percentage of weeds in the 
diet was not affected by animal defoliation during rotational grazing. 

The percentage of dead material in the diet of animals grazing 
pastures under continuous management was higher than that when animals 
sampled rotationally-grazed pastures at the start of the grazing period 
(PC. 10; table 12). No difference in diet dead material percentage was 
found between continuous grazing and after rotational grazing samples 
because dead material percentage increased as an effect of animal 
defoliation during the grazing period (PC. 10). Diet changes were 
similar to the response of dead material to animal defoliation observed 
in the canopy data (table 6) . In the whole canopy, other percentage was 
similar between after rotational grazing and continuous grazing in 
bahia- aeschynomene pastures but higher for continuous grazing in 
bahiagrass pastures (table 4). Canopy layers showed higher dead 
material percentage in continuously-grazed than in rotationally-grazed 
pastures (table 6). In continuously-grazed pastures, the percentage of 
dead material appeared to be much higher in rejected spots of mature 
grass than in the short canopy areas where animals grazed the most. 



59 



Perhaps , animals grazed from areas that contained lower dead material 
percentages than shown for the pasture average. In addition to animal 
selection against dead material, grazing from pasture areas having low 
dead percentage would explain why the effect of grazing management on 
dead material percentage in pastures is not reflected in the diet of the 
animals. An important observation is that the percentages of dead 
material and weeds in the diet of the animals were lower than those 
found in the pasture (tables 6 and 10) , indicating animal discrimination 
against the lowest nutritional components in the pasture. 

Dietary CP concentration was lower for continuous grazing than 
before rotational grazing (P<.10; table 12). No difference in diet CP 
was found between continuous grazing and after rotational grazing. 

Animal defoliation in rotationally- grazed pastures resulted in reduced 
dietary CP concentration (P<.10). Changes in diet CP agreed with those 
observed in the grazing horizon of bahiagrass and bahia-aeschynomene 
pastures (table 8). Higher CP levels in the diet as compared to the 
whole canopy (table 4) and even to the grazing horizon indicate animal 
selectivity. 

Diet IVOMD of animals was higher when grazing pastures under 
rotational management than when grazing pastures under continuous 
management ( P< .10; table 12). A similar response was observed for whole 
canopy forage IVOMD (table 3) and grazing horizon forage IVOMD (table 
8). Animal defoliation during rotational grazing did not affect diet 
IVOMD. In the whole canopy and canopy layers, however, forage IVOMD was 
reduced due to animal defoliation. These results and the higher IVOMD 
levels in the diet as compared to the whole 



canopy and canopy layers 



60 



indicate animal selectivity. The results of diet and pasture canopy CP 
and IVOMD suggest that animals selected for those botanical components 
in the pasture of highest nutritive value. 

Pasture type- grazing management interactions 

There was an interaction between pasture type and grazing 
management on diet grass leaf percentage (P-.001; table 13). A lower 
percentage of grass leaf was found in the diet of animals under 
continuous grazing than under rotational grazing of bahiagrass pastures 
(P<.10; table 13). For bahia-aeschynomene pastures diet grass leaf 
percentage under continuous grazing was higher than that before 
rotational grazing but similar to that after rotational grazing. Animal 
defoliation during rotational grazing reduced diet grass leaf 
percentage in bahiagrass pastures. In bahia-aeschynomene pastures 
defoliation increased diet grass leaf percentage (PC. 10). 

The effects of grazing management and animal defoliation on grass 
leaf percentage in the diet of animals grazing bahiagrass pastures were 
similar to those for grass percentage in the whole canopy (table 4) and 
within the canopy layers (table 7). The magnitude of the differences, 
however, was smaller in the diet. Animals grazing bahiagrass 
rotationally consumed a diet with a grass leaf percentage similar to 
that in the whole canopy. Under continuous grazing, however, animals 
consumed a diet with 40% more grass leaf than that in the canopy. 
Comparing the composition of the diet with the composition of the 
grazing horizon, animals consumed 18, 37, and 78% more grass leaf than 
that in the grazing horizon of bahiagrass pastures at the beginning and 
end of rotational grazing and under continuous grazing, 



61 



T> 

c 

m 



60 

c 

60 -4 
C N 



41 

q- n M 
Cfl 60 
44 14 
60 r4 
0) TO 
U T3 C 
C O 

CO -4 
44 
CO 
44 
O 

<44 < 14 

o oa 



3 

u 

w 

cO 

o.. 



a 

o 



o 

S3 

4J fl} V_/ 

O S 
CO O l-i 
l-i C 10 
<0 4-1 

44 A 4-1 
C O cfl 

•H CD 
<0 X) 

HI <0 s 
A ■ <o 

44 cfl 
-r4 

H A O 

0 co as 
<44 a aa 

w 

01 T) 

CCS 
cfl cfl )-i 
41 O 

e /-v <44 
« 0) 
Cfl V-* A 
0) — / 

14 Cfl 
cfl Cfl 4-> 
3 cfl C 
V U V 
cfl 60 S 

cfl 41 

O *ri 60 
Cfl J3 Cfl 
Cfl Cfl c 
41 A <0 



41 

60 

cfl C 

44 -4 

C 

co a 

O 41 
1-1 44 
41 Cfl 

a 

•o 

e c 

41 cfl 
4-1 

Cfl 14-1 
cfl 

•O 41 

£ i- • 

cd 

0) 

S 

cd 3 

a> bo 
»— 1 0) 

CO 

CO iJ 

cd a) 
n *h 

60 Td 



y C OO 
0) o » 

H C 
a 4i ^ 
u 

C 41 Cfl 
O C4-( 4) 
<4-1 1-1 
41 3 



U 41 91 

U C Cfl 

w 4 ) a 

E 

60 <D <D 
C 60 G 

•H Cfl 41 

N C 
cfl cfl 
E 



E 
O 

5 

60 J3 
C O 

-r4 Cfl 
N 4) 
cfl cfl 
14 • 

60 cfl 
4-1 v4 

G A A 
O c 4) 
O cfl 



14 

60 

Cfl 

3 

o 

3 

C 



XI 

cfl 

H 



cd 




f— 1 


CT\ 




CM 


w 


CU 


o 


ON 


CO 


t— 


o 




o 


O 


O 


co 


•H 




• 


• 


. 


• 


4J 












CO 












*H 












U 




o 


O 


CM 


oo 


cd 


£ 


CNI 


r>* 


CM 


ON 


u 


Cd 


• 


• 


. 




co 


CO 


r-H 


o 


CO 


r— < 






rO 


o 


a 








LO 


o 


m 


00 




o 


CO 


CO 


o 


in 




<3 






4— 1 


t— i 








£} 


o 










00 




ON 




O 


• 


• 




. 


< 








m 


ON 


0Q 










fH 






o 


*0 


X 








co 


oo 


VO 






O 


• 


. 








cd 


r^ 


o 


r^ 


o 




CQ 


CO 




CM 


CM 


















CM 


vO 








O 


• 


• 








u 


CM 


CO 










r** 












o 










o 




f — i 






0Q 


cd 




• 










m 


CO 






















X 










o 


<t 


ON 








cd 




. 








CQ 


CM 


CM 










oo 
















dP 


dP 






dP 


dP 














« 


« 








- 


4-1 


& 






4-1 


s 


cd 


a) 






cd 


0) 


<u 


■u 






a) 


4J 


rH 


CO 








CO 














0) 


a> 






co 


CO 


p 






B 


CO 


CO 


3 






41 


cd 


cd 


60 


60 




4J 






41 


a) 




M 


O 


O 


XI 





4-1 

c 

41 

14 

41 

<44 

<4-1 



41 

14 

cfl 

Cfl 

14 

4) 

4-1 

4-1 

41 



G 

41 

14 

41 41 
3 <4-1 
r-4 <44 
Cfl -r4 

> T3 
>^A 

44 44 



J2 41 

cfl a. 

A 

0 4J 
14 

a. 4i 

1 14 
Cl, 3 

4J 

.. oi 
C Cfl 

cfl a 
41 

E C 

»iH 

<4-4 A 
O 4-1 
•r4 

14 3 
O 

14 3 
14 O 
41 14 

■3 Cfl 
14 

Cfl C 

Tfl -.4 ,-v 

G O 
Cfl 0) ,-4 
4-1 C • 
Cfl cfl V 

II 41 CL, 

X X 
Ed 

913 

O 

cfl J3 



62 



respectively. These results indicate animal selection for grass leaf 
in bahiagrass pastures. Animal selection of grass leaf became more 
intense as the percentage of grass leaf was reduced in the canopy, 
particularly in the grazing horizon. 

In continuously- grazed bahia-aeschynomene pastures, animals 
consumed a diet that was greater in grass leaf percentage compared to 
the diet consumed by cattle at the beginning of the rotational grazing 
period (PC. 10); table 13). Grass leaf percentage in the diet of animals 
increased as an effect of animal defoliation (PC. 10). The response of 
grass leaf percentage to grazing management in bahia-aeschynomene 
pastures is related to the presence of the legume in the canopy and its 
defoliation by animals. Diet grass leaf percentage was somewhat 
similar to grass percentage in the whole canopy (table 4) and grass leaf 
percentage in bottom layer of bahia-aeschynomene pastures (table 7). 
Animals consumed a diet having 63, 64, and 90% more grass leaf than that 
in the grazing horizon (table 8) at the beginning and end of rotational 
grazing and during continuous grazing, respectively, in bahia- 
aeschynomene pastures. These results indicate that relative to the 
whole canopy animals grazing bahia-aeschynomene pastures did not appear 
to select grass leaf. However, animals strongly selected for grass leaf 
when the diet was compared to the grazing horizon, particularly under 
continuous grazing. Brown et al. (1987) reported no indication of 
selection of grass leaf relative to the upper layer of stargrass- 
aeschynomene pastures. 

Grass stem percentage in the diet was affected by grazing 
management in bahia-aeschynomene but not in bahiagrass pastures (P-.099; 



63 



table 13). The dietary levels of this botanical component were low, as 
the percentage of grass stem was also low in the canopy, and probably of 
no biological importance . 

Animals grazing bahia-aeschynomene pastures at the beginning of 
rotational grazing had a greater percentage of legume leaf in the diet 
than did animals under continuous grazing (P<.10; table 13). Legume 
leaf percentage in the diet was reduced by animal defoliation during 
rotational grazing to levels similar to those found under continuous 
grazing (P<.10). Responses of diet legume leaf percentage to grazing 
management and animal defoliation were similar to those observed for 
legume leaf percentage in the whole canopy (table 4) and in the grazing 
horizon (table 6). Diet legume leaf percentage was 86, 42, and 75% 
greater than that in the grazing horizon for animals grazing at the 
beginning and end of rotational grazing and under continuous grazing, 
respectively. These results indicate that animals selected for legume 
leaf. Brown et al. (1987) reported higher levels of dietary legume leaf 
than those in the upper layer at the beginning of the grazing period in 
stargrass-aeschynomene pastures. 

Grazing management and animal defoliation did not affect diet 
legume stem percentage (table 13). Legume stem percentage before 
rotational grazing was 27% lower in the diet than in the grazing 
horizon of bahia-aeschynomene pastures (table 6). After rotational 
grazing animals consumed similar percentages of legume stem as those 
found in the grazing horizon. Animals under continuous grazing 
consumed 32% more legume stem than that in the grazing horizon. 



64 



As a total, animals consumed 48.3, 25.3, and 26.3% legume at the 
beginning and end of rotational grazing and under continuous grazing, 
respectively. Total legume consumed was 118, 113, and 150% greater than 
the percentages of legume present in the whole canopy of these pastures, 
respectively. Relative to the grazing horizon, where most of the legume 
leaf and fine stem were located, under rotational grazing the percentage 
of total dietary legume was 10% greater. Moore and Sollenberger (1986) 
reported diet legume percentages equal to or higher than that in the 
upper layer of limpograss-aeschynomene pastures. Under continuous 
grazing, however, animals consumed a diet containing 46% more legume 
than was present in the grazing horizon. 

The percentage of grass leaf in diet of animals grazing bahiagrass 
pastures indicates that animals selected for this botanical component, 
particularly in relation to the grazing horizon. Animals were not 
forced to select as much under rotational grazing management as they 
were under continuous grazing. The proportion of green material in the 
canopy was probably the driving force to animal selectivity. In 
rotationally-grazed bahia-aeschynomene pastures animals did not have to 
select much for legume as a total because most of the legume available 
in the pasture was at the top of the canopy. In continuously-grazed 
bahia-aeschynomene pastures animals intensified legume selection because 
the legume not only was palatable to them but because in these pastures 
animals were forced to select for green herbage. In all cases animals 
selected for legume leaf as compared to legume stem. Animals selected 
for legume stem when they were forced to do so by the reduction of green 
material, i.e., under continuous grazing. In this case, however, the 



65 



selection for legume leaf was twice as much as for legume stem. The 
presence of legume leaf at the top of the canopy, its palatability and 
selection by the animal did not influence negatively the consumption of 
grass leaf. Grass leaf was better distributed across the canopy than 
was the legume. The greater dietary grass leaf percentage relative to 
that in the grazing horizon indicated that animals consumed the legume 
in the top of the canopy and went down the canopy to consume the grass 
leaf, also. So, in both pasture types, animals selected for green 
material, namely, legume leaf and/or grass leaf. Animals avoided 
consuming dead material, weeds and probably woody legume stem. Because 
the presence of these botanical components, particularly weeds and dead 
material, were in lower percentage when pastures were grazed 
rotationally , it was much easier to harvest desirable green material 
early in rotational grazing than under continuous grazing. 

Cycle effects 

Diet legume leaf percentage changed with cycle (P<.001; table 14). 
The response was similar to the response of legume percentage to cycle 
in the whole pasture canopy (table 5). Even the magnitude of the 
levels of legume leaf percentage in the diet were somewhat similar to 
the whole canopy legume percentage. The legume leaf percentage in the 
diet was higher than the legume leaf percentage in the grazing horizon 
by about 63% and 86% in September and October, respectively (table 9). 
Legume leaf percentage in the grazing horizon (table 9) showed similar 
levels in September and October, as did the diet (table 14). In 
November, legume leaf percentage in the grazing horizon (table 9) and 



66 



Table 14 Least squares means for the effect of cycle on diet 

legume leaf and stem percentage in bahia-aeschynomene 
pastures (n=12) 







Cycle a 




Statistics^ 


Item 


1 


2 3 


4 


SEM 


P 


Legume leaf, % 


14.3 d 


17 . 8 cd 22. l c 


3 . 8 e 


1.99 


.001 


Legume stem, % 


7 . 9 e 


14. 4 d 39. l c 


13. 8 d 


2.03 


.001 



a Cycles correspond to August, September, October and November, 
respectively 

b SEM-standard error of mean; P-probability value 

cde Means in a row with different letters are different (PC. 10) 



67 



legume percentage in the whole canopy (table 5) dropped and so did the 
legume leaf percentage in the diet of the animals. The low availability 
of legume, particularly legume leaf, in November, was related to the 
reproductive stage of the legume plant as mentioned earlier. 

Legume stem percentage in the diet of animals showed response to 
cycle, also (P<.001; table 14). Both legume leaf percentage and legume 
stem percentage in the diet peaked in October as did legume percentage 
in the whole canopy (table 5). Relative to the legume stem percent in 
the grazing horizon (table 9), the diet showed a similar peak in 
October; however, this was still 25% higher than that in the grazing 
horizon. 

The whole pasture canopy and canopy layers showed that the response 
of the legume to season was more a response of the legume stem than of 
the legume leaf. A similar response occurred in the diet of the 
animals; however, animal selectivity was an important factor in the 
higher levels of legume, particularly of legume leaf, in the diet. 

There was an interaction between pasture type and cycle for dietary 
CP concentration (P-.001; table 15). Crude protein in the diet of 
animals grazing bahiagrass was very similar across the season. Animals 
grazing bahia-aeschynomene pastures consumed a diet of increasing CP 
concentration as the season progressed until November when diet CP 
decreased (P<.10). Animals on bahia-aeschynomene pastures had much 
higher levels of CP in the diet than did those in bahiagrass pastures 
PC. 001), with the difference increasing from August to October and then 
declining in November. The response of CP in the diet of animals 
grazing bahia-aeschynomene pastures follows the response of legume 



68 



Table 15 Least squares means for the interaction of cycle and 
pasture type on diet crude protein concentration (n-6) 



Pasture 

type 3 




Cycle' 3 




Statistics 0 


1 


2 


3 


4 


SEM P 


Bahiagrass 


9 . 0 e 


9 7 de 


9 . 5 de 


10. 0 d 


0.37 .001 


Bahia- aeschy. 


12. 0 e 


13. 4 e 


17. 6 d 


12. 5 e 




PDIFF 


.001 


.001 


.001 







a Bahia-aeschy-bahia-aeschynomene pastures; PDIFF-probability of the 
difference 

k Cycles correspond to August, September, October and November, 
respectively 

c SEM-standard error of mean; P-probability value 

Means in a row with different letters are different (P<.10) 



69 



percentage in the diet (table 14) and in the whole canopy (table 5) 
from August to November, and the response of legume stem percentage in 
the canopy layers from September to November (table 9). The levels of 
CP in the diet of animals grazing bahiagrass or bahia-aeschynomene 
pastures were higher than the levels of forage CP found in the whole 
canopy (table 4) or canopy layers. Animal selection of more nutritious 
botanical components is clear from these data, particularly in 
bahiagrass pastures where selection of botanical components with high CP 
concentrations is not as obvious as in bahia-aeschynomene pastures. 

Animal Ingestive Behavior 

Analyses of variance for ingestive behavior data are in table 38, 
appendix. 

Pasture type main effects 

Animals grazing bahiagrass pastures had a larger bite weight than 
did animals on bahia-aeschynomene pastures (P-.012; table 16). Intake 
per minute, however, was similar between pasture types because animals 
on bahia-aeschynomene pastures had a higher number of bites per minute 
than did those on bahiagrass pastures (P-.014). The faster biting rate 
compensated for the smaller bite weight of animals grazing bahia- 
aeschynomene pastures. Brown et al. (1987) reported that cattle grazing 
stargrass pastures had a slower bite rate but larger bite weight 
compared to cattle grazing stargrass-aeschynomene pastures. The 
combination of bite rate and bite weight in the mentioned study, 
however, resulted in greater intake per minute for animals grazing the 
grass-alone pastures. The number of bites per 100 jaw movements 
(B100JM) were not different between animals grazing bahiagrass or 



70 



Table 16 Least squares means for the effect of pasture 
type (bahiagrass (B) and bahia-aeschynomene 
(BA) on animal ingestive behavior (n-24) 



Item a 


Pasture 


Statistics^ 


B 


BA 


SEM 


P 


Bites/min 


40.7 


44.3 


0.74 


.014 


Bites/100 JM 


54.6 


57.5 


1.49 


.223 


Bite weight, g OM 


0.56 


0.50 


0.012 


.012 


Intake /min, g OM 


23.0 


22.6 


0.56 


.658 



a JM— jaw movements; OM— organic matter 
b SEM-standard error of mean; P-probability value 



71 



bahia-aeschynomene pastures. This means that animals did as much 
manipulation in one type of pasture as in the other in order to harvest 
the forage. In stargrass-aeschynomene pastures (Brown et al., 1987), 
cattle tended to manipulate forage less compared to grass-alone 
pastures . 

Grazing management main effects 

The B100JM of animals under continuous grazing was similar to that 
before rotational grazing but higher than that after rotational grazing 
(P<.10; table 17). The B100JM were reduced as an effect of animal 
defoliation during the rotational grazing period. Brown et al. (1987) 
reported reduction in B100JM at the end of the grazing period in 
stargrass-aeschynomene and stargrass pastures. 

Bites per 100 jaw movements is a measurement that can indicate when 
an animal finds it difficult to harvest the most desirable botanical 
components in the pasture. This is because the animal might need to 
make more manipulations with tongue and mouth in order to graze in a 
particular canopy. The more the manipulations, the lower the number of 
actual bites per 100 jaw movements. There was no clear indication in 
the canopy data, however, of the reason why animals on rotationally- 
grazed pastures had more problems to harvest forage at the end of the 
grazing period than did animals on continuously-grazed pastures. None 
of the canopy characteristics were affected by grazing management and 
animal defoliation in the way B100JM was affected. An observation in 
the field was that the fistulated animals appeared to be reluctant to 
graze the pastures at the end of the rotational grazing period. They 
did more walking and smelling of the forage than they normally did in 



72 



Table 17 Least squares means for the effect of grazing 
management (before (BRG) and after (ARG) 
rotational grazing, and continuous grazing (CG)) 
on animal ingestive behavior (n-16) 



Item a 




Management 


Statistics^ 


BRG 


ARG 


CG 


SEM 


P 


Bites/min 


46. 9 C 


36. 9 e 


43. 8 d 


0.91 


.001 


Bites/100 JM 


60. 7 C 


49. 0 d 


58. 4 C 


1.82 


.009 


Bite weight, g OM 


0 . 67 c 


0.48 d 


0 . 45 d 


0.015 


.001 


Intake/min, g OM 


30. 7 C 


17. 6 e 


20. 0 d 


0.68 


.001 



a JM— Jaw movements; OM—organic matter 
k SEM-standard error of mean; P-probability value 
cde Means in a row with different letters are different 
(P<.10) 



73 



the other treatments. They also chewed and kept their heads up for 
longer periods. Apparently the fresh trailing and soiling of the 
pastures had a great impact on grazing behavior of the animals. Then, 
the decreased B100JM in animals after rotational grazing might be due 
more to a reluctance to graze than to an actual difficulty in harvesting 
forage in the pasture. Wade and Le Du (1981) reported that the presence 
of sheep excreta had a greater effect on herbage intake of calves than 
did the spatial distribution of the swards. 

Animals obtained a larger bite weight before rotational grazing 
than under continuous grazing (P< . 10 ; table 17). No difference in bite 
weight was found between the end of rotational grazing and continuous 
grazing because bite weight was reduced as an effect of animal 
defoliation during rotational grazing. The effect of grazing 
management and animal defoliation on bite weight is similar to that on 
whole canopy herbage mass and canopy height (table 3) discussed earlier. 
Changes in bite weight do not appear to be related to grass leaf in the 
whole canopy of either pasture, but they might follow the effect 
observed on whole canopy legume percent and forage CP in bahia- 
aeschynomene pastures (table 4). The response of bite weight to grazing 
management and animal defoliation was also similar to that for forage CP 
and IVOMD in the grazing horizon of both type of pastures (table 8). 
Decreased bite weight due to reduction in herbage mass or leaf material 
has been reported by others (Stobbs, 1973a; Hendricksen and Minson, 

1980; Forbes and Hodgson, 1985; Brown et al., 1987). 

Intake per minute was 50% higher before rotational grazing than 
under continuous grazing (P<.10; table 17). Animal defoliation, 



74 



however, depressed intake per minute by about 43% to levels even lower 
than those found under continuous grazing. This decreased intake per 
minute was due to decreases in both bite weight and bites per minute at 
the end of the rotational grazing period (table 17). Bite weight under 
continuous grazing was similar to that after rotational grazing but 
biting rate was higher under continuous grazing (PC. 10; table 17) 
resulting in a higher intake per minute . 

Pasture type -grazing management interaction 

In bahiagrass pastures the number of bites per minute before 
rotational grazing was similar to that under continuous grazing (table 
18). In bahia-aeschynomene pastures, however, the number of bites per 
minute before rotational grazing was higher than that under continuous 
grazing (PC. 10). In both pasture types, bites per minute were reduced 
as an effect of animal defoliation during rotational grazing (PC. 10). 

The magnitude of the reduction was similar in both cases, about 21%. 
Relative to the continuous grazing system, however, animals on 
bahiagrass pastures after rotational grazing had a biting rate 23% 
lower. No difference in this respect was found in animals grazing 
bahia-aeschynomene pastures. Others have reported increased bites per 
minute with reduction in herbage mass due to defoliation (Allden, 1962; 
Chacon and Stobbs, 1976; Scarnecchia et al., 1985). Chacon and Stobbs 
(1976) reported, however, that further decrease in herbage mass at later 
stages of defoliation reduced not only bite weight but also bites per 
minute . 

These effects of grazing management and animal defoliation on 
biting rate of animals grazing bahia-aeschynomene pastures are similar 



75 



Table 18 Least squares means for the interaction of 
pasture type and grazing management 
(before (BRG) and after (ARG) rotational 
grazing, and continuous grazing (CG)) on 
bites per minute (n-8) 



Pasture 

type 3 


Management 


Statistics* 3 


BRG 


ARG 


CG 


SEM 


P 


Bahiagrass 


43. 4 C 


34. 2 d 


44. 6 C 


1.28 


.033 


Bahia- aeschy . 


50. 5 C 


39. 4 d 


43. 0 d 






PDIFF 


.008 


.027 


.422 







3 Bahia-aeschy-hahia-aeschynomene pastures; PDIFF= 
probability of the difference 
k SEM-standard error of mean; P-probability value 
cd Means in a row with different letters are 
different (P<.10) 



76 



to the effects on whole canopy legume percentage (table 4) and grazing 
horizon legume leaf percentage (table 6) described earlier. In 
bahiagrass pastures, however, the responses of biting rate are not 
matched by any other response observed in the pasture canopy due to 
grazing management and animal defoliation. Early in this discussion, 
the possibility was mentioned that animal trailing and forage soiling 
might affect ingestive behavior of the animals at the end of a 
rotational grazing period. The biting rate data suggest that such an 
effect might be more likely in bahiagrass pastures than in bahia- 
aeschynomene pastures. Apparently, the presence of legume at the end of 
rotational grazing periods overrides the negative effect of trailing and 
soiling °n grazing behavior of animals observed in bahiagrass pastures. 
Bites per minute in rotationally-grazed pastures were greater in bahia- 
aeschynomene than in bahiagrass pastures (PC. 03). When these two 
pasture types were grazed continuously, however, bites per minute were 
similar. Brown et al. (1987) reported greater bite rate in stargrass- 
aeschynomene than in stargrass pastures, and a reduction of bite rate as 
the pastures were defoliated. Moore and Sollenberger (1986) found more 
legume in the diet of animals when ingestive behavior involved increased 
manipulative activity in limpograss-aeschynomene pastures. In the 
present experiment legume percentage was higher when bites per minute 
were higher, i.e., when there was less manipulative activity. 



77 



Pasture Canopy Characteristics and Animal Ingestive 
Behavior Relationships 



Bahiagrass Pastures 

Animal ingestive behavior was affected by only a few 
characteristics of the whole canopy in bahiagrass pastures (table 19). 
Bite weight and intake per minute were positively correlated with canopy 
height. Biting rate was not related to any canopy characteristic in 
bahiagrass pastures. So, as canopy height decreased, bite weight 
decreased and so did intake per minute. Further, animals were not able 
to compensate for smaller bite weight by increasing the number of bites 
per minute. Chacon and Stobbs (1976) have shown bites per minute as a 
compensatory mechanism for decreasing bite weight. In face of a 
decreasing bite weight animals may increase their rate of biting to 
maintain constant intake. Characteristics of the canopy may be reached, 
however, when biting rate cannot increase. Under these conditions of 
small bite weight and limiting biting rate, intake is reduced, as was 
probably the situation in bahiagrass pastures in the present experiment. 
The stubble height allowed in these pastures was probably not 
responsible for the low bite rate because height of the canopy was 
similar after rotational grazing and under continuous grazing. Bites 
per minute, however, were greater under continuous grazing. Therefore, 
something other than canopy characteristics per se affected bites per 
minute at the end of rotational grazing. Grass 

percentage in the whole canopy tended to be negatively related to B100JM 
(table 19). This relationship means that animals tended to make more 
manipulative activities when grass percentage was higher. 



78 



Table 19 Simple correlations 
between whole canopy 
characteristics and 
animal ingestive behavior 
on bahiagrass pastures 
(n-12) 



Item a 


Height 


Grass % 


r b 


pC 


r P 


B100JM 






-.57 .051 


BW 


.56 


.057 




I PM 


.63 


.027 




a B100JM=bites per 


100 jaw 


movements ; 



BW— bite weight; IPM=intake per minute 
r-correlation coefficient 
c P-probability value 



79 



Bite weight was positively correlated with herbage mass in the 
grazing horizon (r-.62, P-.042; n-12) . No other correlation between 
bite weight and canopy layer characteristics was present. The relation 
of bite weight to herbage mass in the top of the canopy and to canopy 
height strongly suggests a larger bite weight in the top canopy. So, a 
reduction in herbage mass in the top layer and a reduction in canopy 
height resulted in a reduction of bite weight. 

Bahia-aeschvnomene Pastures 

Bites per minute were positively correlated with herbage mass, 
canopy height, legume percentage and forage CP in the whole canopy 
(table 20). Jamieson and Hodgson (1979) reported increased bites per 
minute as green herbage mass was progressively reduced. Scarnecchia et 
al. (1985) reported positive correlations between bites per minute and 
plant height in crested wheatgrass pastures. Moore and Sollenberger 
(1986) reported a tendency for increased bites per minute as legume 
percentage was reduced in the canopy of limpograss-aeschynomene 
pastures. In the present study, whole canopy forage CP percentage and 
herbage mass showed the highest correlation with bites per minute. 

Legume percentage (n=12) was highly and positively correlated with 
herbage mass (r-.91, P-.001) and forage CP (r=.88, P-.001). In 
bahiagrass pastures, however, no canopy characteristic was correlated 
with bites per minute. The high legume percentage is associated, then, 
with the higher biting rate observed for animals grazing bahia- 

aeschynomene compared to bahiagrass pastures before rotational grazing 
(table 18). 



80 



cd 

e 

•rA 

C 

cd 

C 

cd 

W CM 
O T-i 
•H U 

c 

tn w 

•H 

p tn 

y y 

P P 
0 3'-' 
y p y 
(-1 tn 3 
to to i-l 
X P. to 
o > 

y 

^ c 
a. y o 
o B *P 
C O -M 
to c -P 
y s^x 
x y 
y o X 
i-i w o 
o y p 
X y cm 
3 i I 

y cp 
c -p 
y jz •- 
y y j-> 
3 x e 
p y 



y 


p 




X 


o 


o 




<44 


•rM 


tn 




4-< 


C 


p 


<4-4 


o 


o 


a) 


•r^ 




o 


V 


> 


o 


cd 


cd 




r— < 


JC 


C 


a> 


a) 


o 


u 


X 3 


•H 


u 




iJ 


o 


<1> 


cd 


a 


> 


»-H 




•H 


0) 


a) 


4J 


>M 


r— J 


w 


>M 


a 


(1) 


o 


6 


bO 


o 


•H 


C 


1 


CO 


•H 


JM 



o 

CM 



XI 

y 

H 







r-< 








CM 




CM 


o 








O 


dP 




o 








O 


O 














Pm 














a 


























ON 




P 


00 








n** 








lO 




r— 4 






CM 




CO 




rH 




dP 






o 




o 


















0) 














rC 














4-> 






»“M 




o 




O 


P 




vO 

i 




i 








in 










dP 


CM 










CM 






o 








o 


<D 




• 










bO 














0) 




00 








CO 


►J 


P 


lO 








VO 






r— 1 


CO 




rH 


r-M 




CM 


o 


o 




o 


o 


4J 

r+ 




o 


O 




o 


o 


bO 




























y 














X 




00 


CM 




r— 1 


CM 




P 




lO 




ON 


ON 






r— ^ 




r—\ 




ON 




CM 


o 




CM 




O 






o 




O 




O 


X 


















CM 




uo 




»— 1 




P 


00 




vO 














£ 








y 






*“3 








E 






o 








y 


£ 


£ 


o 




X 




p 


Pm 


Pm 


rH 




Pm 




M 


CQ 


*-) 


CQ 


CQ 


M 



3 

y 



o 

o 



p 

y 

p. 

« 

y 

4-1 



X 

o 

o 



y 

p 

3 y 
c u 

-4 3 

E c 

•r-4 

P E 
y 

p. p 

y 

tn p, 
p 

c y 
y X 
e y 

y 4-> 
> C 

0 -P 
E » 

3 S 
y m 

1 •- 

S 4-1 

a, X 

•-} bO 

•fH 

.. a) 

y 3 

4J 

3 y 

c p 

•P -P 

s x 
» 

P s 
y cq 
p. 

tn tn 
y 
p 



X 

I 

X 

Pi 

X 



C 

tn • p 
tn y 
y p 
E O 
P 

y p. 
bO 

y y 
X TJ 
P 3 



P 

o 

II 

CM 

U 



y x o 



81 



Jaw movements per minute were positively correlated with canopy 
height and negatively correlated with other percentage in the whole 
canopy of bahia-aeschynomene pastures (table 20). As canopy height was 
reduced, legume leaf disappeared from the canopy and other herbage 
increased, and then jaw movement rate also decreased. In bahiagrass 
pastures, jaw movement rate was not related to any pasture 
characteristic. Moore and Sollenberger (1986) found no canopy 
characteristic in limpograss -aeschynomene pastures related to jaw 
movements . 

Bites per 100 jaw movements were positively correlated with herbage 
mass (table 20). As the pasture was defoliated animals made more 
manipulations before biting. Animals found very contrasting canopy 
characteristics as they defoliated bahia-aeschynomene pastures. They 
went from grazing a canopy with a lot of legume leaf at the beginning of 
rotational grazing to a canopy with a lot of grass and legume stem at 
the end of rotational grazing and under continuous grazing. In 
bahiagrass pastures the canopy was not as contrasting as it was in 
bahia-aeschynomene pastures so changes in manipulative activity did not 
occur. 

Bite weight was positively correlated with canopy height and 
negatively correlated to other percentage in whole canopy of bahia- 
aeschynomene pastures (table 20). Canopy height (n-12) was positively 
correlated with forage CP (r-.82, P-.001) and legume percentage (r-.75; 
P-.005). The correlation between canopy height and forage CP was 
particularly high because of legume leaf in the top of the canopy. As 



82 



the canopy was defoliated, legume leaf, the major botanical component in 
the top canopy, was reduced and so was pasture herbage mass, legume 
percent, forage CP, and bite weight. This relationship can be seen 
simply as a reduction in the bite weight due to a reduction in legume 
leaf in the canopy. When this happened animals reduced biting rate and 
increased manipulations to obtain the remaining legume leaf in the 
canopy or to obtain grass, the new major component in the canopy after 
the defoliation of the legume . 

Intake per minute was positively correlated with the same pasture 
characteristics as was biting rate (table 20). The highest magnitude 
correlations were with canopy height and forage CP indicating the 
influence of legtime leaf on animal intake. 

Herbage mass in both layers of bahia-aeschynomene pastures was 
positively correlated with bites per minute, particularly herbage mass 
in the bottom layer (table 21). Herbage mass in the bottom layer was 
positively correlated with B100JM. These correlations with herbage 
mass in the bottom layer suggest that as the herbage mass was reduced 
and dead material and weeds increased animals increased manipulative 
activity to select for the green leaf and reduced the rate of biting. 

The influence of legume leaf percentage on animal ingestive 
behavior was confirmed in the grazing horizon and bottom layer of bahia- 
aeschynomene pastures (table 21). The correlations were positive and 
particularly high between legume leaf percentage in the grazing horizon 
and bite weight and intake per minute. Moore and Sollenberger (1986) 
did not find any effect of legume percentage in the upper layer of 
limpograss-aeschynomene pastures on bite weight. Legume leaf percentage 



83 



Table 21 Simple correlations between canopy layer 

characteristics and animal ingestive behavior on bahia- 
aeschynomene pastures (n-12; r-correlation coefficient; 
P-probability value) 



Item a 






HM C 


LLd 


% 


Dead % 


Weeds % 


L b 


r 


P 


r 


P 


r P 


r P 


BPM 


2 


.55 


.081 


.63 


.036 




-.53 .097 




1 


.73 


.007 


.55 


.063 




-.60 .040 


B100JM 


2 
















1 


.76 


.004 










BW 


2 






.88 


,001 


-.71 .015 






1 






.59 


043 






I PM 


2 






.85 


001 


-.65 .032 






1 






.59 


042 







a BPM— bites per minute; B100JM=bites per 100 jaw movements; 
BW-bite weight; IPM-intake per minute 

L-layer (2-grazing horizon, over 20 cm above ground level; 
1-bottom layer, 10-20 cm above ground level) 
c HM-herbage mass 
^ LL-legume leaf 



84 



in both layers was also positively correlated with bites per minute. 
These correlations confirm the positive influence of legume leaf on 
biting rate observed in the ingestive behavior data discussed earlier. 

Influences of dead material and weeds on ingestive behavior were 
also observed. Dead material in the grazing horizon was negatively 
correlated with bite weight and intake per minute (table 21). As with 
the canopy data (table 6), changes in dead material due to defoliation 
might occur faster in the grazing horizon than in the bottom layer and 
explain why there are correlations of dead material with bite weight and 
intake per minute for grazing horizon but not for the bottom layer. When 
the increase in dead material occurs in the grazing horizon, the animal 
probably becomes more selective for green material. As a consequence, 
bite size and intake per minute are reduced. Moore and Sollenberger 
(1986) observed lower bite weight and lower intake per minute when 
pastures of limpograss-aeschynomene were shorter and when there was a 
high proportion of dead material in the upper layer. Both studies 
su gg est that animals actively avoided dead herbage present in the 
grazing horizon. 

Weed percentage was negatively correlated with bites per minute in 
both layers (table 21). This influence did not have much effect on 
intake per minute . Perhaps dead material more than weeds has a 
negative influence on performance of animals grazing bahia- 
aeschynomene pastures. 

It is evident from these results that legume and particularly 
legume leaf is the botanical component in bahia-aeschynomene pastures 
for which the animals selected. The presence of legume gave a 



85 



contrasting canopy structure and nutritive value to the pasture. The 
animal searched for more nutritious components in the canopy by making 
almost full use of its ingestive behavior mechanisms. Moore et al. 
(1985) suggested that dietary legume percentage of animals grazing 
limpograss-aeschynomene pastures was affected by different factors than 
those which affect total intake. 

In bahiagrass pastures the canopy was relative consistent in its 
components and nutritive value. Consequently, animals were also quite 
consistent in their ingestive behavior. Only small changes in their 
ingestive behavior were required to maintain maximum intake of herbage 
available in the pasture. 



SUMMARY AND CONCLUSIONS 



Introducing the legume aeschynomene into bahiagrass pastures 
increased herbage mass in the pastures. Aeschynomene also increased 
forage CP in the canopy grazing horizon by about 40%. The whole canopy 
forage IVOMD was depressed slightly (6%) by the introduction of the 
legume. Animals selected for a higher nutritive value diet in both 
pasture types. Animals obtained a 46% higher CP and only 6% lower IVOMD 
diet when grazing bahia- aeschynomene than when grazing bahiagrass 
pastures . 

Bite weight obtained by animals grazing bahia -aeschynomene pastures 
was lower than that obtained from bahiagrass pastures. Intake per 
minute was not different, however, because in legume -grass pastures 
there was a higher biting rate which may have compensated for smaller 
bite weight. Bahia-aeschynomene pastures in the present study had an 
average of 15% legume in the canopy. The responses of the variables 
measured in the present experiment probably would change depending upon 
the percentage of legume in the pasture. 

Herbage mass in bahiagrass and bahia- aeschynomene pastures was 
increased by managing the pastures in a rotational rather than in a 
continuous grazing system when both rotational and continuous pastures 
were grazed to 20 cm. Animal defoliation during rotational grazing 
reduced herbage mass to levels similar to those in continuously- grazed 
pastures . 



86 



87 



Forage IVOMD in bahiagrass and bahia-aeschynomene pastures was also 
increased by managing the pastures rotationally . Even though animal 
defoliation reduced forage IVOMD during the rotational grazing period, 
the levels were still higher than those under continuous grazing. Part 
of this response was due to a reduction in the accumulation of dead 
material in the canopy of rotationally- grazed pastures. 

By grazing the pasture rotationally animals consumed less weeds. 
Dead material was also consumed in lower percentage at the beginning of 
rotational grazing compared to that during continuous grazing. 

Animals were able to consume a higher CP diet at the beginning of 
the rotational grazing period than when they were under continuous 
grazing. Animals consumed a higher IVOMD diet when under rotational 
grazing than when under continuous grazing. 

Rotational grazing allowed animals to obtain a large bite weight at 
the beginning of the grazing period. This advantage, however, 
disappeared at the end of the grazing period when levels were similar to 
those found under continuous grazing. Due to the positive relationship 
between bite weight and intake per minute, animals at the beginning of 
the rotational grazing period obtained a high intake per minute. A 
depression in the rate of biting of animals after rotational grazing, 
however, depressed intake per minute to levels even lower than those 
found under continuous grazing. The data indicated that the presence of 
legume in bahia-aeschynomene pastures may override some of the 
depression in biting rate at the end of rotational grazing periods. 
Depression in bite rate at the end of rotational grazing periods may 



88 



have been due to an effect of the fresh trailing and soiling of the 
forage during the grazing period. 

Also, managing the pastures rotationally allowed a higher 
percentage of grass in bahiagrass pastures and a higher accumulation of 
legume in bahia-aeschynomene pastures. The higher legume percentage in 
bahia-aeschynomene pastures was accompanied by a higher forage CP in the 
pasture at the beginning of the rotational grazing period. 

In conclusion, the data indicated that the introduction of the 
legume aeschynomene in a bahiagrass pasture might increase the herbage 
mass available and the nutritive value of the forage for s umm er grazing 
animals. Managing bahiagrass and, particularly, bahia-aeschynomene 
pastures in a rotational grazing system favored a more desirable canopy 
structure and forage nutritive value for a higher diet nutritive value 
and intake. The more desirable canopy characteristics, however, might 
disappear at the end of rotational grazing periods with a negative 
effect on animal response. 

A careful grazing management strategy is suggested if bahia- 
aeschynomene pastures are to be managed in a rotational grazing system. 
More efficient animals or animals of higher nutritional requirements 
should graze first during the rotational grazing period. It might not 
be a benefit at all if these animals are allowed to graze the pastures 
during the entire period. Animals will have the opportunity for a 
higher intake of a forage of higher nutritive value before all the 
legume is defoliated from the pasture. So, maximum benefit will be 
obtained if animals are removed to a new paddock before total 
defoliation of the legume. Other animals of lower nutritional needs 



89 



might graze the remaining forage. A biological response is foreseen if 
bahiagrass and particularly bahia-aeschynomene pastures are grazed 
rotationally rather than continuously. The more complicated management, 
however, deserves economical considerations that were not among 
objectives of the present study. Type of production system and 
marketing characteristics of a specific location surely have an impact 
on the economical returns of more complicated grazing systems. 

The Florida livestock industry has calves and lactating cows, 
efficient animals in terms of weight gain, that might benefit as first 
grazers of the grazing system. Strategies of how such a grazing system 
would fit within the whole management system of a ranch need to be 
designed or investigated. There is a potential benefit to the beef 
industry in Florida in the use of this grass -legume association if 
management is done as indicated by the results of the present 
experiment. Uncertainty of such a system, like in many other 
agricultural activities, has to be expected if weather conditions are 
not adequate to obtain a good stand of aeschynomene . 



APPENDIX 



91 









rH 


rH 


rH 






rH 


rH 


CM 
















&4 


o 


o 


VO 






o 


o 


oo 


CM 
















o 


o 


O 






o 


o 


Ox 


O 












O 




• 


• 


• 






• 


• 


• 














H 
































a) 






























*H 


X 






























cd 


x 






























o 


o 






CO 


oo 




CM 


CM 


rH 


ON 


CO 




On 






•h 






• 


• 






• 


• 


. 




• 










C 




CO 


m 


On 


m 






vO 


co 


CM 


in 




VO 






cd 




x 


<t 


<r 


CM 




CM 


O 


in 




m 




rH 






4J 






rH 


o 


rH 






O 


rH 














O 






rH 


CM 








CM 
















XI 
































'd 
































c 






rH 


rH 


rH 






rH 


rH 


O 












cd 




d4 


O 


o 


O 






o 


ON 


m 


ON 










U 






o 


o 


o 






o 


O 


ON 


o 










X 


C/l 






























bO 


w 






























•H 


cd 






























0) 


x 






























X 


o 




vO 


CM 


m 




CM 


ON 


o\ 




oo 




o 






« 






00 


in 


co 




•n 


00 


CM 


m 


in 




rH 






w 




CO 


VO 


m 


on 




rH 


CM 


m 








CM 






C/l 




x 


CM 


rH 


CO 






m 
















d a) 

a 3 








rH 








CM 
















r— 1 
































<U tfl 
00 > 


4-> 




CM 


iH 


00 






rH 


CM 


rH 


CM 










tfl 


X 


o> 


VO 


o 


CM 






o 


rH 


o 


O 










X 


b0 






o 


co 






o 


o 


o 


rH 










V4 4J 


•H 




• 


• 


. 






. 






. 










0) -r4 


(0 






























X i— 1 


X 






























•H 
































































a « 


a 




CM 


co 


00 




o 


oo 


CO 














2 


o 


CO 


vO 


n* 


CO 




CM 


CO 


in 


in 


CM 




CM 






C 0 


c 


X 


• 


• 






« 


• 












tB >-i 


ca 




O 


in 


00 




vO 


CO 




ON 


co 




rH 






o cu 


u 






o 








in 
















1 








VO 
























<0 C4 
rH 
































o •- 
,c a> 






VO 


CM 


<* 






rH 


rH 


rH 


rH 










S 4i 


tfl 


a. 


O 


o 


CM 






o 


o 


o 


o 










tfl 


to 




o 


o 


rH 






o 


o 


o 


o 










M 3 


tfl 




• 


• 


• 






















o tr 


a 






























4-1 U5 


<0 






























<u c 


oo 




























rH 


O tB 


tfl 




<t 


CM 


CM 




rH 


CO 




rH 


m 




oo 




tfl 


C <0 


X 




on 


rH 


oo 




<t 


in 


CO 




o 








-r4 


tfl e 






<fr 


rH 


CO 




On 


rH 


VO 


CO 


in 




o 






i 


0) 


CO 


in 


VO 


CO 




CO 


CO 




CO 


00 




CO 




d) 


1-4 CO 


X 


X 




CM 


m 




00 


00 


o 


o 


vO 




CM 




4J 


tfl 52 








VO 


CM 






CM 


CM 


CO 


rH 






tfl 


> 






rH 


rH 






















a 


44 C 
0 o 




XI 


























T3 

tfl 

a tu 

O TJ 
Tfl 


•H 
tfl 4-> 
•H »H 




d.f 


rH 


CM 


CM 




vO 


CO 


CO 


vO 


v£> 




oo 


r>. 


C/1 C/1 

>x O 
rH Q* 

cd g 

c 0 

< O 








/-s 

X 


















rH 




<u n 
0) C 

<U V4 tfl 
4J 44 
tfl to 

O 44 Tfl 






tfl 

c 

o 






















u 




•r4 O 0) 


CNI 


44 

O 


/^\ 


X 

e 

a; 






X 

* 


pH 

o 










0) 

Td 

c 




.-1 0) 
a to 3 
<u to m 
u <u n 
1 U4 <u 
Q- t»0 X 


CM 

a> 


0) 

o 


•H 

4J 

cd 


H 


S 

Cl) 

b0 




cd 

u 


H 

a 


a) 






X 

* 


X 

u 


•H 

cd 

B 


»H 


rH 

X 

cd 


V4 

3 

o 


•H 

u 

cd 


d) 

a 


cd 

C 

cd 


X 

* 


o 

u 

u 


<d 

o4 


rH 

o 

>*> 


H 

* 

>* 


X 

* 


H 

* 

>4 


o 

u 

u 


0) 

04 


cd 

X 

o 


(U <U 4J 
OS Q O 


H 


CO 


> 


H 


X 


H 


w 




u 


O 


O 


U 


w 




H 


tfl jQ O 



92 



Table 23 Analysis of variance for 
whole canopy legume 
percentage in bahia- 
aeschynomene pastures 
(MS=mean square ; 
P-probability value) 



Source of 
variation 3 




Legume 


d.f , b 


MS 


P 


Management (M) 


2 


331.1 


081 


Error a 
Rep(M) 


3 


51.0 




Cycle (CY) 


3 


309.5 


001 


CY*M 


6 


5.4 


586 


Error b 
Remainder 


9 


6.7 




Total 


23 







3 Rep=replicate 
b Degress of freedom 



93 



Table 24 Analysis of variance for whole canopy 
nutritive value (MS-mean square; 
P=probability value) 



Source of 




Crude 


protein 


IV0MD c 


variation 3 


d. 


f. b MS 


P 


MS 


P 


Type (T) 


1 


16.08 


.012 


112.8 


.054 


Management (M) 


2 


11.73 


.015 


155.9 


.021 


T*M 


2 


5.04 


.080 


22.8 


.376 


Error a 
Rep(T*M) 


6 


1.27 




19.7 




Cycle (CY) 


3 


1.33 


.036 


243.3 


.001 


CY*T 


3 


7.68 


.001 


9.9 


.383 


CY*M 


6 


0.66 


.167 


11.0 


.355 


CY*T*M 


6 


2.72 


.001 


9.9 


.416 


Error b 
Remainder 


18 


0.38 




9.2 




Total 


47 











3 Rep-replicate 
b Degrees of freedom 

c IVOMD-in vitro organic matter digestion 



94 



Table 25 Analysis of variance for canopy layer herbage mass, and 

botanical composition (MS=mean square; P=probability value) 



Source of 




Herbage 


mass 


Grass leaf 




Grass 


stem 


variation 3 


d.f. b 


MS 


P 


MS 


P 


MS 


P 


Type (T) 


1 


1148688 


.001 


11855.7 


.001 


45.98 


.124 


Management (M) 


2 


171882 


.044 


1584.7 


.001 


7.47 


.620 


T*M 


2 


187133 


.038 


863.1 


.004 


7.06 


.635 


Error a 
Rep(T*M) 


6 


31396 




53.0 




14.39 




Layer (L) 


1 


4278807 


.001 


1699.3 


.001 


32.89 


.144 


T*L 


1 


249284 


.005 


484.9 


.001 


30.99 


.154 


M*L 


2 


8351 


.565 


14.1 


.245 


5.15 


.662 


T*M*L 


2 


1786 


.877 


97.8 


.007 


4.82 


.679 


Error b 

Rep*L(T*M) 


6 


13272 




7.8 




11.65 




Cycle (CY) 


2 


76126 


.016 


2508.3 


.001 


39.56 


.062 


CY*T 


2 


43900 


.078 


255.2 


,011 


37.31 


.071 


CY*M 


4 


18122 


.348 


50.2 


,392 


7.00 


.697 


CY*T*M 


4 


11598 


.568 


67.3 


252 


6.64 


.717 


CY*L 


2 


12378 


.461 


62.0 


284 


27.61 


.134 


CY*T*L 


2 


5519 


.704 


133.0 


078 


26.02 


.149 


CY*M*L 


4 


10109 


.630 


64.6 


270 


5.19 


.799 


CY*T*M*L 


4 


18092 


.349 


26.5 


690 


4.89 


.815 


Error c 
Remainder 


24 


15463 




46.8 




12.61 




Total 


71 















* Rep-replicate; only cycles 2, 3, and 4 are included 
b Degress of freedom 



O' to 



95 



Table 26 Analysis of variance for canopy layer 
botanical composition (MS=mean 
square; P=probability value) 



Source of 




Weeds 




Dead material 


variation 3 


d.f. b 


MS 


P 


MS 


P 


Type (T) 


1 


315.3 


.144 


76.9 


.215 


Management (M) 


2 


97.2 


.465 


2551.9 


.001 


T*M 


2 


13.7 


.886 


146.8 


.091 


Error a 
Rep(T*M) 


6 


111.6 




39.9 




Layer (L) 


1 


563.2 


.003 


348.0 


.001 


T*L 


1 


2.4 


.759 


1.6 


.707 


M*L 


2 


2.3 


.908 


94.4 


.015 


T*M*L 


2 


18.7 


.496 


2.3 


.804 


Error b 
Rep*L(T*M) 


6 


23.6 




10.2 




Cycle (CY) 


2 


60.8 


.113 


2403.2 


.001 


CY*T 


2 


282.1 


.001 


161.8 


.007 


CY*M 


4 


221.0 , 


.001 


92.8 


.021 


CY*T*M 


4 


70.8 


.050 


70.6 


.055 


CY*L 


2 


14.5 . 


,574 


100.8 


.035 


CY*T*L 


2 


146.0 . 


009 


0.33 


.987 


CY*M*L 


4 


82.2 . 


030 


27.8 


.340 


cy*t*m*l 


4 


17.1 . 


618 


10.5 


.805 


Error c 
Remainder 


24 


25.4 




26.1 




Total 


71 











Rep-replicate; only cycles 2, 3, and 4 are included 
Degress of freedom 



96 



Table 27 Analysis of variance for canopy layer legume 
fractions, (MS=mean square; P-probability 
value) 



Source of 




Le gume 


leaf 


Legume 


stem 


variation 3 


d.f. b 


MS 


P 


MS 


P 


Management (M) 


2 


120.4 


.030 


374.8 


.016 


Error a 
Rep(M) 


3 


8.6 




16.9 




Layer (L) 


1 


357.5 


.001 


580.5 


.012 


M*L 


2 


85.5 


.002 


89.5 


.125 


Error b 
Rep*L(M) 


3 


1.0 




19.9 




Cycle (CY) 


2 


143.8 


.001 


668.0 


.001 


CY*M 


4 


17.6 


.037 


30.0 


.223 


CY*L 


2 


54.5 


.002 


167.8 


.004 


CY*M*L 


4 


14.1 


.068 


5.2 


.883 


Error c 
Remainder 


12 


4.9 




18.2 




Total 


35 











* Rep-replicate; Only cycles 2, 3, and 4 are included 
° Degress of freedom 



ns x 



97 



Table 28 Analysis of variance for canopy layer nutritive 
value (MS=mean square; P-probability value) 



Source of 
variation 3 


d.f , l 


Crude protein 


IV0MD c 


5 MS 


P 


MS 


P 


Type (T) 


1 


71.78 


.001 


128.91 


.076 


Management (M) 


2 


26.42 


.006 


322.06 


.009 


T*M 


2 


4.70 


.177 


31.24 


.388 


Error a 












Rep(T*M) 


6 


2.01 




28.06 




Layer (L) 


1 


55.42 


.001 


6.99 


.124 


T*L 


1 


19.58 


.001 


11.23 


.064 


M*L 


2 


8.38 


.001 


28.95 


.006 


T*M*L 


2 


2.95 


.016 


8.70 


.079 


Error b 












Rep*L(T*M) 


6 


0.33 




2.18 




Cycle (CY) 


2 


3.34 


.002 


416.00 


.001 


CY*T 


2 


16.26 


.001 


34.75 


.016 


CY*M 


4 


1.70 


.011 


28.34 


.012 


CY*T*M 


4 


2.73 


.001 


19.4 


.051 


CY*L 


2 


1.60 


.034 


6.92 


.389 


CY*T*L 


2 


1.47 


.043 


6.80 


.395 


CY*M*L 


4 


0.17 


.795 


4.44 


.645 


CY*T*M*L 


4 


0.30 


.585 


1.49 


.930 


Error c 












Remainder 


24 


0.41 




7.04 




Total 


71 











Rep-replicate 
Degress of freedom 
c IVOMD-in vitro organic matter digestion 



Analysis of variance for grazing horizon herbage mass and botanical 
composition (MS— mean square; P— probability) 



98 







co 


00 


CM 




CM 


rH 


CM 


00 








Put 


CO 


r- 


ax 




ax 


o 


o 












CM 




ax 




rH 


o 


o 


CM 






W 
















































a> 
























a> 
























& 




tH 


in 


in 


vo 


CM 


ax 




CM 


ax 






co 


• 


• 


• 


• 


• 


• 


• 


• 


. 






X 


tH 


CM 


o 






vo 


CO 


ax 


co 








CO 


VO 




r-. 


VO 


tH 


00 




CO 








rH 










<t 


CM 












r*. 


O 


vo 




rH 


co 


00 


vo 






e 


pu. 


CO 




m 




tH 


CM 




vo 






a> 




tH 


VO 


vo 




tH 


tH 










u 




• 


• 


• 




• 


. 


. 


. 






CO 
























CO 




CM 


in 


oo 


ax 


VO 


00 


tH 


m 


tH 




CO 




• 


• 


• 


• 


• 


• 


• 


• 


• 




cd 


CO 


VO 


CM 


tH 


m 


vo 


CM 


CM 


tH 


m 






X 


r* 


tH 


tH 


CM 


vo 


vo 


tH 


tH 


CM 




O 




























rH 




<t 




tH 


CO 




00 






4-1 


Put 


o 


o 


o 




o 


CM 


'tf 


ax 






cd 




o 


o 


o 




o 


o 


CM 


m 






Q) 




• 


• 






• 


. 


. 


. 






rH 
























CO 




o 


CM 




vo 


00 


oo 


O 


co 






CO 




• 


• 


• 


• 


• 


. 




. 






cd 




00 


CM 




o 




Mt 


co 


tH 


tH 






CO 


VO 


m 


m 




m 




tH 


in 






o 


X 


m 


vo 


vo 




VO 


co 


tH 












00 








rH 


















rH 


m 




CM 


ax 




vo 






co 


PLt 


o 


CM 


o 




CO 






CM 






CO 

/■H 




o 


O 


o 




tH 


tH 




oo 






vw 

6 
























a) 
























bO 
























cd 




ax 


m 


m 


vO 


o 


CM 


m 


m 


in 




xi 


C/5 


vo 


CM 


ax 


CO 


oo 


in 


oo 


CO 


00 




h 


X 


00 


ax 


ax 


CO 


m 


vO 


oo 


m 


vo 




a> 




CO 


m 




oo 


CM 


ax 




vo 


r^* 




X 




vo 


vo 


o 






CO 






tH 








tH 




tH 










































44 
























• 


rH 


CM 


CM 


vo 


CM 


CM 






CM 


m 






















tH 


co 
































2 
























■'w' 






































u 






cd 




4J 














<1) 




4-1 


C 




a 




2 


>* 












o 


o 


/~N 


(1) 




* 


b 








c 






•H 


H 


E 




cd H 


V-/ 








X* *H 




a> 


■U 


V-' 


0) 




V-/ 








52 


cd 




o 


cd 




bO 




^ o. 


a> 






-X 


U B 


rH 


V-J 


*H 


a) 


cd 




o <u 


rH 


H 


2 


H 


o a) 


cd 


d 


u 


a 


C 


2 


^ 05 


o 


* 


* 


■X 


U (X, 


■U 


o 


cd 




cd 


* 


1-1 








>h 


u 


o 


cn 


> 


H 


2 


H 


tu 


o 


CJ 


cj 


o 


w 


H 



efl £) 



Rep-replicate; Only cycles 2, 3, and 4 are included 
Degress of freedom 



99 



Table 30 Analysis of variance for grazing horizon dead material and 
forage nutritive value (MS-mean square; P— probability 
value ) 



Source of 
variation 3 


d.f. 


Dead material 


Crude protein 


IVOMD c 


b MS 


P 


MS 


P 


MS 


P 


Type (T) 


1 


28.2 


.292 


83.17 


.001 


32.0 


.134 


Management (M) 


i 2 


1756.0 


.001 


32.28 


.002 


260.5 


.001 


T*M 


2 


65.0 


.121 


6.73 


.063 


9.4 


.467 


Error a 
















Rep(T*M) 


6 


21.2 




1.48 




10.8 




Cycle (CY) 


2 


1678.9 


.001 


4.74 


.001 


262.1 


.001 


CY*T 


2 


86.3 


.108 


12.18 


.001 


29.3 


.030 


CY*M 


4 


73.8 


.118 


1.38 


.034 


18.2 


.066 


CY*T*M 


4 


36.5 


.384 


1.72 


.017 


12.1 


.164 


Error b 
















Remainder 


12 


32.0 




0.37 




6.2 




Total 


35 















* Rep-replicate; Only cycles 2, 3, and 4 are included 
Degrees of freedom 

c IVOMD-in vitro organic matter digestion 



100 



Table 31 Analysis of variance for grazing horizon 
legume fractions (MS=mean square ; 
P=probability value) 



Source of 




Legume 


leaf 


Legume 


stem 


variation 3 


d.f , b 


MS 


P 


MS 


P 


Management (M) 


2 


202.0 


.010 


411.5 


.019 


Error a 
Rep(M) 


3 


6.6 




21.3 




Cycle (CY) 


2 


184.9 


.002 


723.2 


.002 


CY*M 


4 


30.3 


.082 


27.2 


.566 


Error b 
Remainder 


6 


8.9 




33.9 




Total 


17 











* Rep=replicate ; Only cycle 2, 3, and 4 are included 
b Degrees of freedom 



Analysis of variance for bottom layer herbage mass and botanical composition 
(MS-mean square; P-probability value) 



101 







o 


co 


VO 




CM 


oo 


on 


CM 










p-t 


CO 


r^ 


rH 






co 




CM 












rH 


in 


VO 




in 


m 


co 


rH 








w 


















* 


































Q) 


























<D 






























VO 


rH 


00 


VO 


o 


rH 


vo 






o 






C/3 


• 


• 


• 


• 


• 


• 


. 


. 










£ 


VO 




rH 


o 


rH 


rH 


On 


oo 












00 


CO 


co 


VO 


rH 


rH 


rH 


CO 




rH 








rH 


























eg 


on 


o 




rH 


CJv 


ON 


VO 








s 


P-i 


CO 




VO 




VO 


VO 


rH 


CO 








0) 

jj 




o 


CO 


CO 




O 


O 












w 


























co 


























w 






CM 


CM 


On 




rH 


oo 


oo 




m 




cd 


CO 




rH 


rH 


O 


m 


in 


o 


o 




rH 




Jh 


X 


• 


• 


• 


• 


. 


. 












O 




o 


o 


o 


o 


o 


o 


o 


o 




o 








«h 


rH 


m 




rH 


00 


vo 


00 








C4-I 


CL, 


o 


o 


o 




O 


in 


00 


vo 








cd 

d) 




o 


o 


o 




O 


in 


ON 


rH 








rH 


























CO 




VO 


m 


tH 


CM 


vO 




00 


m 




ON 




CO 




• 


• 


• 


• 


. 




• 










cd 


CO 


CM 


VO 


CO 


O 


m 


co 


rH 


CM 




rH 




u 


E 


r-* 


<1" 


o 


CM 


rH 


rH 








CM 




o 






On 


CO 




on 


















CO 


























«-h 


vO 


o 




m 


ON 












CO 


Ph 


o 


rH 


00 




vO 


On 


m 


o 








CO 

cd 




o 


rH 


«H 




o 




CM 


CM 








6 


























a) 


























b0 




co 


00 


CO 


CO 






vo 


Ht 




rH 




cd 




o 


o 


CM 


CO 


CM 


VO 




m 










CO 


rH 


CO 


ON 


CO 


ov 




CO 


rH 




CM 






£ 






CO 


VO 


m 


On 


o 


CO 




CO 




a) 




co 


rH 


00 


CO 






CM 


CM 




rH 




X 




CM 


rH 
























rH 


















































<4-1 


























T3 


rH 1 


CM 


CM 


VO 


CM 


CM 








CM 


in 
























rH 


co 








/^s 


























X 
















































cd 




4J 
















U 

d) 






C 




c 




X 


>* 










T3 




O 


o 


/~s 


0) 




* 


O 










p 






•H 


H 


s 




Cd H 


'w' 










•H 




<D 


•U 




0) 




'w' 












cd 




O 


cd 




oo 




u a 


a) 






* 


u 


S 


rH 


>H 


•H 


a) 


cd 




o <v 


rH 


H 


X 


H 


o 


a) 


cd 


d 




Q* 


G 


X 


Jh 06 


O 


* 


* 


* 


u 


Dh 


u 


O 


cd 




cd 


* 


U 




>* 


>H 




u 




o 


CO 


> 


H 


s: 


H 


W 


o 


O 


u 


O 


w 




H 



(0 ,£) 



Rep— replicate ; Only cycles 2, 3, and 4 are included 
Degress of freedom 



102 



Table 33 Analysis of variance for bottom layer dead material and 

forage nutritive value (MS-mean square; P-probability value) 



Source of 
variation 3 


d.f. 


Dead material 


Crude protein 


IV0MD c 


a MS 


P 


MS 


P 


MS 


P 


Type (T) 


1 


50.2 


.236 


8.19 


.021 


108.1 


.056 


Management (M) 


i 2 


890.2 


.001 


2.52 


.128 


90.5 


.060 


T*M 


2 


84.2 


.131 


0.92 


.398 


30.5 


.282 


Error a 
















Rep(T*M) 


6 


28.9 




0.85 




19.4 




Cycle (CY) 


2 


825.1 


.001 


0.20 


.654 


160.9 


.001 


CY*T 


2 


75.8 


.054 


5.56 


.001 


12.2 


.253 


CY*M 


4 


46.9 


.116 


0.49 


.406 


14.6 


.185 


CY*T*M 


4 


44.5 


.130 


1.31 


.067 


8.8 


.400 


Error b 
















Remainder 


12 


20.2 




0.45 




7.9 




Total 


35 















* Rep-replicate; Only cycles 2, 3, and 4 are included 
b Degrees of freedom 

c IVOMD-in vitro organic matter digestion 



103 



Table 34 Analysis of variance for bottom layer 
legume fractions (MS=mean square ; 
P-probability value) 



Source of 




Legume leaf 


Legume 


stem 


variation 3 


d.f , b 


MS 


P 


MS 


P 


Management (M) 


2 


3.94 


.396 


52.8 


.169 


Error a 
Rep(M) 


3 


3.07 




15.5 




Cycle (CY) 


2 


13.49 


.009 


112.5 


.001 


CY*M 


4 


1.47 


.379 


7.9 


.098 


Error b 
Remainder 


6 


1.16 




2.5 




Total 


17 










a Rep-=replicate ; 


Only cycles 2, 3, 


and 4 


are included 



b Degrees of freedom 



Analysis of variance for diet botanical composition (MS— mean square; 
P— probability value) 



104 



cO 






ro 


vO 




rH 


ro 


r— l 


ON 




•t-i 


Ou 


uo 


O 


ro 




o 


in 


o 


H 




u 




o 


O 


On 




o 


o 


o 


o 




<D 




• 


• 


• 




. 


. 








4J 






















<0 






















e 




CM 




ro 




00 


vO 


ro 


oo 


ro 


T3 




• 


• 


• 


• 


• 


• 


. 




<0 


CO 


ro 


<* 


rH 


ON 




on 


00 


rH 


VO 


<U 


X 


rH 


ro 




rH 


CM 


H 


in 


CM 


Q 




rH 


ro 






ro 













CM 


in 


ro 




rH 


CM 


rH 


in 




Oh 


rH 


VO 


r— 1 




o 


O 


CM 


rH 






o 


o 


ro 




o 


o 


O 


o 


Cfl 




• 


• 


• 




. 








T3 




















Q) 




















<D 

3 




ro 


ro 


00 


vO 


in 






ON 


CO 


o 


ON 






ON 


CM 


On 


rH 




X 


rH 


o 


ro 


CM 


rH 


vO 


CM 


ro 






ro 


rH 






rH 









a 

<D 

4J 

W 

W 

W 

cd 

U 

O 



<4H 

(0 

0) 



W 

CO 

cO 

J-i 

o 



co 

£ 



T3 



On 

in 



rH 


ON 




rH 


rH 


CM 


CM 


vO 


ON 




o 


o 




CM 


O 


O 




o 


o 


ro 


O 


ro 


00 


o 


CM 


r^ 


ON 


CM 


00 


ro 




CM 


ro 


rH 


m 


rH 


rH 




ON 


rH 





rH UO rH 

o co o 

O CM o 



co 

VO 

VO 



CM 

rH 

co 



CM 



o 

o 



o 

o 



rH co 

rH o 

^ 00 



rO 

O 

o 



o 

ro 



vO 



vO 



vO 



m 

CM 






6 

o 

T3 

0) 

0) 



4h 


cd 




4J 






/~s 








U 

Q) 

'O 




G 




c 




X 


>* 










O 


o 


/~s 


0) 




■X 










f2 




<1> 


•H 


H 


a 




m h 


V-/ 








n >h 




■U 


V-/ 


a> 




'w' 








52 


cd 

^ 6 
o a> 

Jh 




O 

U 

3 


cd 

•H 

u 


<D 

a 


00 

cd 

c 


X 


u a 
o <u 
m o4 


0) 

rH 

O 


H 

■X 


X 

* 


* 

H 

* 


rH 

cd 

4_) 


o 

CO 


cd 

> 


>0 

H 


cd 

£ 


* 

H 


u 

u 


CJ 


>* 

o 


>1 

u 


O 


>H 

W 


O 

H 



CO £1 



Rep-replicate 



105 



Table 36 Analysis of variance for diet legume 
fractions of animals grazing bahia- 
aeschynomene pastures (MS=mean square; 
P-probability value) 



Source of 




Legume 


leaf 


Legume 


stem 


variation 3 


d.f , b 


MS 


P 


MS 


P 


Management (M) 


2 


1078.9 


.034 


55.1 


.312 


Error a 
Rep(M) 


3 


83.4 




31.3 




Cycle (CY) 


3 


365.5 


.001 


1151.7 


.001 


CY*M 


6 


163.9 


.006 


39.4 


.253 


Error b 
Remainder 


9 


23.8 




24.7 




Total 


23 











a Rep=replicate 
15 Degrees of freedom 



106 



Table 37 Analysis of variance for diet nutritive value 
(MS-mean square; P-probability value) 



Source of 




Crude 


protein 


IV0MD c 




variation 3 


d.f , b 


MS 


P 


MS 


P 


Type (T) 


1 


225.1 


.003 


221.9 


.003 


Management (M) 


2 


84.3 


.015 


47.3 


.048 


T*M 


2 


21.8 


.171 


26.4 


.129 


Error a 
Rep(T*M) 


6 


9.1 




9.0 




Cycle (CY) 


3 


20.4 


.001 


74.5 


.003 


CY*T 


3 


19.6 


.001 


9.8 


,478 


CY*M 


6 


5.6 


.001 


12.2 


413 


CY*T*M 


6 


5.9 


.001 


40.7 


016 


Error b 
Remainder 


18 


0.82 




11.3 




Total 


47 











a Rep-replicate 
b Degrees of freedom 

c IVOMD-in vitro organic matter digestion 



107 









00 


tH 


oo 




VO 


CM 


Ox 


in 








c 


Cm 


in 


o 






VO 


CM 




CM 






/—N 


•H 




VO 


o 


co 




O 


O 


O 


O 






0) 


e 




• 


• 


• 
















3 


\ 
























rH 


0) 
























cd 


X 
























> 


cd 




VO 


vt 


m 


m 


in 


rH 


in 


CM 




rv 




■U 


C/S 


• 


• 


• 


• 


. 


. 


. 


. 




. 




C 


X 


rH 


CM 


o\ 




o 




oo 


vf 




o 


u 


M 






00 






CO 


<± 


CM 


co 




rH 


•H 








rv 


















rH 


























*H 


























rO 


























cd 




























0) 




CM 


rH 


o 




00 


Ov 


oo 


vf 






o 


N 


PM 


tH 


o 


rH 




o 


m 


o 


co 








•H 




o 


o 


m 




o 


o 


o 


rH 






cx 

H 

CL, 


CO 




• 


• 


• 










* 






a> 








CO 


<t 




















m 


CM 


o 


o 


CM 


rH 


CM 


rH 




o 


• •> 


•H 


CO 


o 


CM 


o 


o 


O 


o 


O 


o 




o 


d> 

U 

cd 

3 


CQ 


a 


o 


O 


o 


o 


O 


o 


O 


o 




o 


cr 


























CO 


























c 


























cd 


as 




CO 


ON 






rH 


rH 


CM 


rv 






a) 


•“> 


PM 


CNI 


o 


CM 




o 


o 


o 


vo 






e 


o 

O 




CM 


o 






o 


O 


o 


o 






C/3 


rH 
























22 


\ 












rH 


00 


CO 


CM 




vo 


V-/ 


co 






CO 


CM 


CM 


• 




• 


. 




. 




<D 




• 


• 


• 


• 


rH 


o 


rv 


CM 




rH 




4J 


CO 


00 


m 


CM 


CO 






CM 


in 




CM 


0 


•H 


X 


Ov 


rH 


m 


in 


m 


rH 


rH 








•H 


PQ 






VO 


















> 


























cd 


























X 


























0) 




















































d> 






CO 


ON 


rH 




CO 


in 


OV 


rv 






> 


C 


PM 


rH 


CO 


ON 




o 


oo 


CM 


rH 






•H 


•H 




CM 


rH 


rH 




o 


CO 


o 


CO 






4J 


e 




• 


• 


• 






. 










CO 


\ 
























0) 


























bO 

G 


o 

22 




vf 


VO 


VO 


o 


rH 


rv 


CM 


vf 




O 


•H 


CO 


o 


CM 


rv 


VO 


o 


o 


00 


oo 




00 






S 


m 




m 


CM 


VO 




rH 


>3- 




CO 


rH 

cd 

e 














CM 




rH 








•H 


























c 


























cd 


























u 








rH 


CO 




rH 


rH 


co 


rH 






o 


C2 


PM 


T— 1 


o 


CO 




O 


o 


o 


o 






4-1 


•H 

6 

\ 




o 


o 


O 




o 


o 


o 


o 






0) 


















* 






o 


CO 
























c 

cd 


<D 

U 




m 




co 


rH 


CO 


m 


VO 


CO 




VO 


•H 


•H 


CO 


m 


co 


co 


CO 


a\ 




Ox 


m 




m 




CQ 


X 


m 


CM 


00 


rH 


rH 




CM 






cd 






rH 


vf 






rH 












> 


























4-1 


























O 




X> 






















CO 




4-1 






















•H 

CO 

>o 




•o' 


rH 


CM 


CM 


VO 


co 


CO 


VO 


vo 




18 


rH 


























cd 








/>> 


















C 








22 


















<2 




























4H 

O 


(fl 




•U 




/•“V 


/-s 










Remainder 


00 


c 

o 


X-N 


C 

<D 




22 

* 


>< 

o 










Table 3 


0) 

o 

d 

o 

C/3 


•H 

U 

cd 

•H 

cd 

> 


Type (T 


e 

0) 

b0 

cd 

c 

cd 

22 


T*M 


Error a 
Rep(T 


'w' 

0) 

rH 

o 

O 


CY*T 


CY*M 


22 

* 

H 

* 

>4 

O 


Error b 



r*"* 

vi- 



ed 

•u 

o 

H 



cfl ^ O 



Rep=replicate 
Degrees of freedom 
JM— jaw movements 



LITERATURE CITED 



Albrecht, S. L. , J. M. Bennett, and K. H. Quesenberry. 1981. Growth 
and nitrogen fixation of Aeschvnomene under water stressed 
conditions. Plant Soil 60:309-315. 

Allden, W. G. 1962. Rate of herbage intake and grazing time in 
relation to herbage availability. Proc. Aust. Soc. Anim. Prod. 
4:163-171. 

AOAC. Association of Official Analytical Chemists. 1975. Official 
Methods of Analysis. (12th Ed.). Washington, DC. 

Archibald, K. A. E. , R. C. Campling and W. Holmes. 1975. Milk 

production and herbage intake of dairy cows kept on a leader and 
follower grazing system. Anim. Prod. 21:147-156. 

Baker, F. S. 1980. Let's produce slaughter beef in Florida. Proc. 
29th Beef Cattle Short Course, pp. 5-7. Animal Sci. Dept., IFAS , 
Univ. of Fla., Gainesville. 

Beaty, R. E. , R. A. McCreery, and J. D. Powell. 1960. Response of 

Pensacola bahiagrass to nitrogen fertilization. Agron. J 52*453- 
455. 

Beaty, E. R. , R. L. Stanley, and J. Powell. 1968. Effect of height of 
cut on yield of Pensacola bahiagrass. Agron. J. 60:356-358. 

Bertrand, J. E. , and L. S. Dunavin. 1988. Grazing evaluation of warm- 
season perennial pasture grasses with beef steers. Florida Beef 
Cattle Research Report, pp. 30-32. IFAS, Univ. of Fla., 
Gainesville . 

Blaser , R. E. 1982. Integrated pasture and animal management. Trop. 
Grassl. 9:24 

Blaser, R. E., R. C. Hammes , J. P. Fontenot, H. T. Bryant, C. E. Polan, 
D. D. Wolf, F. S. McClaugherty , R. G. Kline, and J. S. Moore. 1986. 
Forage-animal management systems. Editor M. C. Holliman. Virginia 
Agr. Exp. Sta. Virginia Polytechnic Institute and State University 
Blacksburg. Bull. 86-7. 

Blue, W. G. 1983. Effects of lime and fertilizer on forage production. 
Proc. 32nd Beef Cattle Short Course, pp. 94-100. Animal Sci. 

Dept., IFAS, Univ. of Fla., Gainesville. 



108 



109 



Brown, W. F., and P. Mislevy. 1988. Influence of maturity and season 
on the yield and quality of tropical grasses. Florida Beef Cattle 
Research Report, pp. 46-54. IFAS . Univ. of Fla. , Gainesville. 

Brown, W. F. , J. E. Moore, and P. Mislevy. 1987. Managing the plant- 
animal interface in tropical grass -legume pastures. Proc. 43d 
Southern Pasture and Forage Crop Improvement Conference, pp. 45-48. 

Chacon, E. , and T. H. Stobbs. 1976. Influence of progressive 

defoliation of a grass sward on the eating behavior of cattle. 

Aust. J. Agric. Res. 27:709-727. 

Chacon, E. A., T. H. Stobbs, and M. B. Dale. 1978. Influence of sward 
characteristics on grazing behavior and growth of Hereford steers 
grazing tropical grass pastures. Aust. J. Agric. Res. 29:89-102. 

Chambliss, C. G. , and D. W. Jones. 1981. Bahiagrass. Fla. Coop. Ext. 
Serv. Circ. 321B. Univ. of Fla., Gainesville. 

Chaparro, C. J. 1989. Management factors affecting Aeschvnomene 

americana L. seed production and natural reestablishment. Master 
Sci. Thesis. Univ. of Fla., Gainesville. 

Forbes, T. D. A., and J. Hodgson. 1985. Comparative studies of the 

influence of sward conditions on the ingestive behavior of cows and 
sheep. Grass and Forage Sci. 40:69-77. 

Hendricksen, R. , and J. Minson. 1980. The feed intake and grazing 
behavior of cattle grazing a crop of Lab lab purnureus cv. Rongai. 

J. Agric. Sci., Camb. 95:547-554. 

Hendricksen, R. E., and D. J. Minson. 1985. Growth, canopy structure 
and chemical composition of Lab lab purnureus cv. Rongai at Sanford, 
S.E. Queensland. Trop. Grassl. 19:81-86. 

Hodges, E. M. , A. E. Kretschmer, P. Mislevy, R. D. Roush, 0. C. Ruelke, 
and G. H. Snyder. 1982. Production and utilization of the tropical 
legume aeschynomene . Circular S-290. IFAS. Univ. of Fla., 
Gainesville . 

Hodges, E. M., F. M. Peacock, H. L. Chapman, and R. E. L. Greene. 1974. 
Forage and supplement systems for beef cows in South-Central 
Florida. Proc. Soil Crop Sci. Soc. Fla. 33:56-58. 

Hodges, E. M., F. M. Peacock, H. L. Chapman, and F. G. Martin. 1976. 
Grazing studies with Pensacola bahiagrass and warm season annual 
legumes. Proc. Soil Crop Sci. Soc. Fla. 36:173-175. 

Hodgson, J. 1982. Influence of sward characteristics on diet selection 
and herbage intake by the grazing animal. In J. B. Hacker (Ed.) 
Nutritional Limits to Animal Production from Pastures. pp. 153-166. 
Commonwealth Agric. Bureaux, Farnham Royal, U.K. 



110 



Hodgson, J. 1985. The control of herbage Intake in the grazing 
ruminant. Proc. Nut. Soc. 44:339-346. 

Jamieson, W. S., and J. Hodgson. 1979. The effects of variation in 

sward characteristics upon the ingestive behavior and herbage intake 
of calves and lambs under a continuous stocking management. Grass 
Forage Sci. 34:273-282. 

Kalmbacher, R. S., and F. G. Martin. 1983. Light penetrating a 
bahiagrass canopy and its influence on establishing Jointvetch. 
Agron. J. 75:465-468. 

Kalmbacher, R. S., P. Mislevy, F. G. Martin, J. W. Prevatt, C. G. 

Chambliss, and G. Kidder. 1988. Establishment of aeschynomene in 
bahiagrass sod. Circular S-355. IFAS, Univ. of Fla., Gainesville 

Koger, M. , W. G. Blue, G. B. Killinger, R. E. L. Greene, H. C. Harris, 

J. M. Myers, A. C. Warnick and N. Gammon Jr. 1983a. Beef 
production, soil and forage analyses, and economic returns from 
eight pasture programs in North-Central Florida. Florida. Bull. 

631 (technical). IFAS. Univ. of Fla., Gainesville. 

Koger, M. , W. G. Blue, G. B. Killinger, R. E. L. Greene, J. M. Myers, N. 
Gammon Jr, A. C. Warnick and J. R. Crockett. 1983b. Production 
response and economic returns from five pasture programs in North- 
Central Florida. Bull. 740 (technical). IFAS. Univ. of Fla., 
Gainesville . 

Kretschmer, A. E. , G. H. Snyder, and T. C. Wilson. 1986. Productivity 
and persistence of selected Aeschynomene spp. Soil Crop Sci. Soc. 
Fla. Proc. 45:174-178 

Kunkle , W. E. , and J. A. Baldwin. 1988. Effect of a protein supplement 
on the performance of calves grazing bahiagrass. Florida Beef 
Cattle Research Report, pp. 95-96. IFAS, Univ. of Fla., 

Gainesville . 

Kunkle, W. E. , D. B. Bates, D. L. Pitzer, and A. C. Hammond. 1988. 

Effect of protein and blackstrap molasses supplements on performance 
of calves grazing bahiagrass pasture. Florida Beef Cattle Research 
Report, pp. 87-94 IFAS, Univ. of Fla. , Gainesville. 

Ludlow, M. M. , T. H. Stobbs, R. Davis, and D. A. Charles -Edwards. 1982. 
Effect of sward structure of two tropical grasses with contrasting 
canopies on light distribution, net photosynthesis and size of bite 
harvested by grazing cattle. Aust. J. Agric. Res. 33:187-201. 

Milford, R. , and K. P. Haydock. 1965. The nutritive value of protein 
in subtropical pasture species grown in south-east Queensland. 

Aust. J. Exp. Agric. and An. Husb. 5:13. 



Ill 



Milford, R. , and D. J. Minson. 1965. Intake of tropical pasture 

species. Proc. IX Int. Grassl. Cong., Sao Paulo, Brazil. pp. 815- 
822. 

Mislevy, P. , and F. G. Martin. 1985. Changes in production and quality 
with plant elongation and maturity of american jointvetch. Proc. XV 
Int. Cong., pp. 1051-1052. Kyoto, Japan. 

Moore, J. E. 1981. Principles of forage quality evaluation, pp. 66- 
87. In King Visiting Scholar Lectures. Ark. Agric. Exp. Sta. 
Special Rep. 93. 

Moore, J. E., and G. 0. Mott. 1974. Recovery of residual organic 

matter from in vitro digestion of forages. J. Dairy Sci. 57:1258- 
1259. 

Moore, J. E. , and L. E. Sollenberger . 1986. Canopy structure effects 

on ingestive behavior. Proc. 42nd Southern Pasture and Forage Crop 
Improvement Conference, pp. 53-57. 

Moore, J. E., G. L. Ellis, C. E. Rios, and M. Koger. 1969. Estimates 
of the voluntary intake and nutrient digestibility of bahiagrass 
pasture by grazing steers. Soil Crop Sci. Soc. Fla. Proc. 29:168- 
175. 

Moore, J. E. , L. E. Sollenberger, G. A. Morantes, and P. T. Beede . 

1985. Canopy structure of Aeschvnomene americana - Hemarthria 
altissima pastures and ingestive behavior of cattle. XV Int. Gras. 
Cong., Kyoto, Japan. 

Ocumpaugh, W. R. 1978. Warm-season forage for maximum beef gains. 

Proc. 27th Beef Cattle Short Course. An. Sci. Dept., IFAS, Univ.of 
Fla. , Gainesville 

Peacock, F. M. , E. M. Hodges, R. E. L. Greene, W. G. Kirk, M. Koger, and 
J. R. Crockett. 1976. Forage systems, beef production and economic 
evaluations, South-Central Florida. Bull. 783 (technical). IFAS, 
Univ. of Fla., Gainesville. 

Pitman, W. D. 1986. Grazing evaluation of tropical legumes. Proc. 37th 
Beef Cattle Short Course, pp. 55-58. IFAS, Animal Sci. Dept. 

Univ. of Fla., Gainesville. 

Pitman, W. D. , and A. E. Kretschmer. 1984. Persistence of selected 
tropical pasture legumes in peninsular Florida. Agron. J. 76:993- 
996. 

Pitzer, D. L. , W. E. Kunkle, and D. B. Bates. 1988. Blackstrap 

molasses and grain supplements fed to heifers grazing bahiagrass 
pasture. Florida Beef Cattle Research Report, pp. 80-86. IFAS. 
Univ. of Fla., Gainesville. 



112 



Prates, E. R. , H. L. Chapman, E. M. Hodges, and J . E. Moore. 1974. 
Animal performance by steers grazing Pensacola bahiagrass pastures 
in relation to forage production, forage composition, and estimated 
intake. Soil Crop Sci. Soc. Fla. Proc. 34:152-156. 

Riewe, M. E. 1980. Designing rational experiments at the plant-animal 
interface: adapted from Dr. H. L. Lucas. Proc. 37th Southern 
Pasture Forage Crop Improvement Conference. Nashville, Tennessee. 

Rusland, G. A., L. E. Sollenberger , K. A. Albrecht, C. S. Jones and L. 

V. Crowder. 1988. Animal performance on limpograss-aeschynomene 
and nitrogen-fertilized limpograss pastures. Agron. J. 80:957-962. 

Sampaio, E. V. S. B. , E. R. Beaty, and D. A. Ashley. 1976. Bahiagrass 
regrowth and physiological aging. J. Range Managmt. 29:316-371 

SAS, 1987. SAS/STAT Guide for personal computers, version 6 edition. 
Cary, NC:SAS Institute Inc., 1028 pp. 

Scarnecchia, D. L. , A. Nastis, and J. C. Malecheck. 1985. Effects of 
forage availability on grazing behavior of heifers. J. Range Mangt 
38:177-180. 

Snedecor, G. W. , and W. G. Cochran. 1980. Statistical methods. 

Seventh edition. Iowa State Univ. Press. Ames, Iowa. 

Sollenberger, L. E. , J. E. Moore, K. H. Quesenberry, and P. T. Beede. 
1987a. Relationships between canopy botanical composition and diet 
selection in aeschynomene- limpograss pastures. Agron. J. 79*1049- 
1054. 

Sollenberger, L. E. , K. H. Quesenberry and J. E. Moore. 1987b. Forage 
quality responses of an Aeschynomene -limpograss association to 
grazing management. Agron. J. 79:83-89. 

Sollenberger, L. E. , W. R. Ocumpaugh, V. P. B. Euclides, J. E. Moore, K. 
H. Quesenberry, and C. S. Jones Jr. 1988. Animal performance on 
continuously stocked 'Pensacola' bahiagrass and 'Floralta' 
limpograss pastures. J. Prod. Agric. 3:216-220. 

Spreen, T. H. , J. A. Ross, J. W. Pheasant, J. E. Moore, and W. E. 

Kunkle . 1985. A simulation model for backgrounding feeder cattle 

in Florida. Bull. 850 (technical). IFAS , Univ. of Fla. , 
Gainesville . 

Stephenson, W. J., and D. E. McCloud. 1987. Ona 1986 climatological 
report. Research report RC-1987-1. IFAS, Univ. of Fla., 

Gainesville . 

Stobbs, T. H. 1973a. The effect of plant structure on the intake of 
tropical pastures. I. Variation in the bite size of grazing cattle. 
Aust. J. Agric. Res. 24:809-819. 



113 



Stobbs, T.H. 1973b. The effect of plant structure on the intake of 
tropical pastures. II. Differences in sward structure, nutritive 
value, and bite size of animals grazing Setaria anceps and Chloris 
gayana at various stages of growth. Aust. J. Agric. Res. 24:821- 
829. 

Stobbs, T.H. 1975. The effect of plant structure on the intake of 

tropical pasture. III. Influence of fertilizer nitrogen on the size 
of bite harvested by Jersey cows grazing Setaria anceps cv. 

Kazungula swards. Aust. J. Agric. Res. 26:997-1007. 

Tilley, J. M. A., and R. A. Terry. 1963. A two-stage technique for the 
in vitro digestion of forage crops. J. British Grassl. Soc. 18:104 

Wade, M. H. , and Y. L. R. Le Du. 1981. Influence of sward structure 
upon herbage intake of cattle grazing a perennial ryegrass sward. 
Proc. XIV Int. Grassl. Cong., Lexington, Kentucky. 

Wheeler, J. L. 1962. Experimentation in grazing management. A review. 
Herbage Abstracts. 32:1-7. 



BIOGRAPHICAL SKETCH 



Jorge Luis Morales G. was born on July 30, 1951, in Cartago, Costa 
Rica. He graduated from Enrique Menzel High School, Turrialba, Costa 
Rica in 1969. In 1976 he started his professional studies at the 
University of Chihuahua, Mexico, and he graduated as a zootechnician 
engineer in 1979. He is married to Graciela Jimenez and has three 
children, Jorge, Diana and Jose Alonso. 

He has been working with the Ministry of Agriculture and Livestock 
of Costa Rica since 1980. He was trained and worked as invited 
researcher in tropical forages in the Centro Internacional de 
Agricultura Tropical (CIAT) , Palmira, Colombia in 1982. He received his 
Master of Science degree in dairy cattle nutrition from Dairy Science 
Department of the University of Florida in 1986. In summer of 1986, he 
started his doctoral program at the Animal Science Department of the 
University of Florida. He is presently a candidate for the Doctor of 
Philosophy degree in animal nutrition. 



114 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is fully 
adequate, in scope and quality, as a dissertation for the degree of 
Doctor of Philosophy. 




<j°, 



Tohn E. Moore, Chairman 
/Professor of Animal Science 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is fully 
adequate, in scope and quality, as a dissertation for the degree of 
Doctor of Philosophy. 






William F. Brown, Co-Chairman 
Associate Professor of Animal Science 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is fully 
adequate, in scope and quality, as a dissertation for the degree of 
Doctor of Philosophy. 








E. Sollenberger 
Assistant Professor of Agronomy 



I certify that I have read this study and that in ray opinion it 
conforms to acceptable standards of scholarly presentation and is fully 
adequate, in scope and quality, as a dissertation for the degree of 
Doctor of Philosophy. 




Harold H. Van Horn 
Professor of Dairy Science 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is fully 
adequate, in scope and quality, as a dissertation for the degree of 
Doctor of Philosophy. 




William G. Blue 
Professor of Soil Science 



This dissertation was submitted to the Graduate Faculty of the 
College of Agriculture and to the Graduate School and was accepted as 
partial fulfillment of the requirements for the degree of Doctor of 
Philosophy. 



August 1990 



ollege of Agrie 



Dean, College of Agriculture 



Am 



Dean, Graduate School